webscout 1.3.6__py3-none-any.whl → 1.3.9__py3-none-any.whl

webscout/AI.py

@@ -20,21 +20,24 @@ import json
 import yaml
 from webscout.AIutel import Optimizers
 from webscout.AIutel import Conversation
-from webscout.AIutel import AwesomePrompts
-from webscout.AIbase import Provider
+from webscout.AIutel import AwesomePrompts, sanitize_stream
+from webscout.AIbase import  Provider, AsyncProvider
 from Helpingai_T2 import Perplexity
-from typing import Any
+from webscout import exceptions
+from typing import Any, AsyncGenerator
 import logging
-#-----------------------------------------------Cohere--------------------------------------------
-class Cohere(Provider):
+import httpx
+#-----------------------------------------------llama 3-------------------------------------------
+class LLAMA2(Provider):
     def __init__(
         self,
-        api_key: str,
         is_conversation: bool = True,
-        max_tokens: int = 600,
-        model: str = "command-r-plus",
-        temperature: float = 0.7,
-        system_prompt: str = "You are helpful AI",
+        max_tokens: int = 800,
+        temperature: float = 0.75,
+        presence_penalty: int = 0,
+        frequency_penalty: int = 0,
+        top_p: float = 0.9,
+        model: str = "meta/meta-llama-3-70b-instruct",
         timeout: int = 30,
         intro: str = None,
         filepath: str = None,
@@ -42,20 +45,17 @@ class Cohere(Provider):
         proxies: dict = {},
         history_offset: int = 10250,
         act: str = None,
-        top_k: int = -1,
-        top_p: float = 0.999,
     ):
-        """Initializes Cohere
+        """Instantiates LLAMA2
 
         Args:
-            api_key (str): Cohere API key.
             is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True.
-            max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 600.
-            model (str, optional): Model to use for generating text. Defaults to "command-r-plus".
-            temperature (float, optional): Diversity of the generated text. Higher values produce more diverse outputs.
-                            Defaults to 0.7.
-            system_prompt (str, optional): A system_prompt or context to set the style or tone of the generated text. 
-                            Defaults to "You are helpful AI".
+            max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 800.
+            temperature (float, optional): Charge of the generated text's randomness. Defaults to 0.75.
+            presence_penalty (int, optional): Chances of topic being repeated. Defaults to 0.
+            frequency_penalty (int, optional): Chances of word being repeated. Defaults to 0.
+            top_p (float, optional): Sampling threshold during inference time. Defaults to 0.9.
+            model (str, optional): LLM model name. Defaults to "meta/llama-2-70b-chat".
             timeout (int, optional): Http request timeout. Defaults to 30.
             intro (str, optional): Conversation introductory prompt. Defaults to None.
             filepath (str, optional): Path to file containing conversation history. Defaults to None.
@@ -64,20 +64,22 @@ class Cohere(Provider):
             history_offset (int, optional): Limit conversation history to this number of last texts. Defaults to 10250.
             act (str|int, optional): Awesome prompt key or index. (Used as intro). Defaults to None.
         """
-        self.session = requests.Session()
         self.is_conversation = is_conversation
         self.max_tokens_to_sample = max_tokens
-        self.api_key = api_key
         self.model = model
         self.temperature = temperature
-        self.system_prompt = system_prompt
-        self.chat_endpoint = "https://production.api.os.cohere.ai/coral/v1/chat"
+        self.presence_penalty = presence_penalty
+        self.frequency_penalty = frequency_penalty
+        self.top_p = top_p
+        self.chat_endpoint = "https://www.llama2.ai/api"
         self.stream_chunk_size = 64
         self.timeout = timeout
         self.last_response = {}
         self.headers = {
             "Content-Type": "application/json",
-            "Authorization": f"Bearer {self.api_key}",
+            "Referer": "https://www.llama2.ai/",
+            "Content-Type": "text/plain;charset=UTF-8",
+            "Origin": "https://www.llama2.ai",
         }
 
         self.__available_optimizers = (
@@ -119,7 +121,7 @@ class Cohere(Provider):
            dict : {}
         ```json
         {
-           "text" : "How may I assist you today?"
+           "text" : "How may I help you today?"
         }
         ```
         """
@@ -134,30 +136,43 @@ class Cohere(Provider):
                     f"Optimizer is not one of {self.__available_optimizers}"
                 )
         self.session.headers.update(self.headers)
+
         payload = {
-            "message": conversation_prompt,
+            "prompt": f"{conversation_prompt}<s>[INST] {prompt} [/INST]",
             "model": self.model,
+            "systemPrompt": "You are a helpful assistant.",
             "temperature": self.temperature,
-            "preamble": self.system_prompt,
+            "topP": self.top_p,
+            "maxTokens": self.max_tokens_to_sample,
+            "image": None,
+            "audio": None,
         }
 
         def for_stream():
             response = self.session.post(
                 self.chat_endpoint, json=payload, stream=True, timeout=self.timeout
             )
-            if not response.ok:
-                raise Exception(
-                    f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
+            if (
+                not response.ok
+                or not response.headers.get("Content-Type")
+                == "text/plain; charset=utf-8"
+            ):
+                raise exceptions.FailedToGenerateResponseError(
+                    f"Failed to generate response - ({response.status_code}, {response.reason})"
                 )
 
+            message_load: str = ""
             for value in response.iter_lines(
                 decode_unicode=True,
+                delimiter="\n",
                 chunk_size=self.stream_chunk_size,
             ):
                 try:
-                    resp = json.loads(value.strip().split("\n")[-1])
-                    self.last_response.update(resp)
-                    yield value if raw else resp
+                    if bool(value.strip()):
+                        message_load += value + "\n"
+                        resp: dict = dict(text=message_load)
+                        yield value if raw else resp
+                        self.last_response.update(resp)
                 except json.decoder.JSONDecodeError:
                     pass
             self.conversation.update_chat_history(
@@ -165,7 +180,6 @@ class Cohere(Provider):
             )
 
         def for_non_stream():
-            # let's make use of stream
             for _ in for_stream():
                 pass
             return self.last_response
@@ -211,20 +225,23 @@ class Cohere(Provider):
         """Retrieves message only from response
 
         Args:
-            response (dict): Response generated by `self.ask`
+            response (str): Response generated by `self.ask`
 
         Returns:
             str: Message extracted
         """
         assert isinstance(response, dict), "Response should be of dict data-type only"
-        return response["result"]["chatStreamEndEvent"]["response"]["text"]
-#-----------------------------------------------REKA-----------------------------------------------
-class REKA(Provider):
+        return response["text"]
+class AsyncLLAMA2(AsyncProvider):
     def __init__(
         self,
-        api_key: str,
         is_conversation: bool = True,
-        max_tokens: int = 600,
+        max_tokens: int = 800,
+        temperature: float = 0.75,
+        presence_penalty: int = 0,
+        frequency_penalty: int = 0,
+        top_p: float = 0.9,
+        model: str = "meta/meta-llama-3-70b-instruct",
         timeout: int = 30,
         intro: str = None,
         filepath: str = None,
@@ -232,16 +249,17 @@ class REKA(Provider):
         proxies: dict = {},
         history_offset: int = 10250,
         act: str = None,
-        model: str = "reka-core",
-        system_prompt: str = "Be Helpful and Friendly. Keep your response straightforward, short and concise",
-        use_search_engine: bool = False,
-        use_code_interpreter: bool = False,
     ):
-        """Instantiates REKA
+        """Instantiates LLAMA2
 
         Args:
-            is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True
-            max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 600.
+            is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True.
+            max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 800.
+            temperature (float, optional): Charge of the generated text's randomness. Defaults to 0.75.
+            presence_penalty (int, optional): Chances of topic being repeated. Defaults to 0.
+            frequency_penalty (int, optional): Chances of word being repeated. Defaults to 0.
+            top_p (float, optional): Sampling threshold during inference time. Defaults to 0.9.
+            model (str, optional): LLM model name. Defaults to "meta/llama-2-70b-chat".
             timeout (int, optional): Http request timeout. Defaults to 30.
             intro (str, optional): Conversation introductory prompt. Defaults to None.
             filepath (str, optional): Path to file containing conversation history. Defaults to None.
@@ -249,25 +267,23 @@ class REKA(Provider):
             proxies (dict, optional): Http request proxies. Defaults to {}.
             history_offset (int, optional): Limit conversation history to this number of last texts. Defaults to 10250.
             act (str|int, optional): Awesome prompt key or index. (Used as intro). Defaults to None.
-            model (str, optional): REKA model name. Defaults to "reka-core".
-            system_prompt (str, optional): System prompt for REKA. Defaults to "Be Helpful and Friendly. Keep your response straightforward, short and concise".
-            use_search_engine (bool, optional): Whether to use the search engine. Defaults to False.
-            use_code_interpreter (bool, optional): Whether to use the code interpreter. Defaults to False.
         """
-        self.session = requests.Session()
         self.is_conversation = is_conversation
         self.max_tokens_to_sample = max_tokens
-        self.api_endpoint = "https://chat.reka.ai/api/chat"
+        self.model = model
+        self.temperature = temperature
+        self.presence_penalty = presence_penalty
+        self.frequency_penalty = frequency_penalty
+        self.top_p = top_p
+        self.chat_endpoint = "https://www.llama2.ai/api"
         self.stream_chunk_size = 64
         self.timeout = timeout
         self.last_response = {}
-        self.model = model
-        self.system_prompt = system_prompt
-        self.use_search_engine = use_search_engine
-        self.use_code_interpreter = use_code_interpreter
-        self.access_token = api_key
         self.headers = {
-            "Authorization": f"Bearer {self.access_token}",
+            "Content-Type": "application/json",
+            "Referer": "https://www.llama2.ai/",
+            "Content-Type": "text/plain;charset=UTF-8",
+            "Origin": "https://www.llama2.ai",
         }
 
         self.__available_optimizers = (
@@ -275,7 +291,6 @@ class REKA(Provider):
             for method in dir(Optimizers)
             if callable(getattr(Optimizers, method)) and not method.startswith("__")
         )
-        self.session.headers.update(self.headers)
         Conversation.intro = (
             AwesomePrompts().get_act(
                 act, raise_not_found=True, default=None, case_insensitive=True
@@ -287,17 +302,20 @@ class REKA(Provider):
             is_conversation, self.max_tokens_to_sample, filepath, update_file
         )
         self.conversation.history_offset = history_offset
-        self.session.proxies = proxies
+        self.session = httpx.AsyncClient(
+            headers=self.headers,
+            proxies=proxies,
+        )
 
-    def ask(
+    async def ask(
         self,
         prompt: str,
         stream: bool = False,
         raw: bool = False,
         optimizer: str = None,
         conversationally: bool = False,
-    ) -> dict:
-        """Chat with AI
+    ) -> dict | AsyncGenerator:
+        """Chat with AI asynchronously.
 
         Args:
             prompt (str): Prompt to be send.
@@ -306,10 +324,10 @@ class REKA(Provider):
             optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
             conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
         Returns:
-           dict : {}
+           dict|AsyncGeneraror[dict] : ai content
         ```json
         {
-           "text" : "How may I assist you today?"
+           "text" : "How may I help you today?"
         }
         ```
         """
@@ -324,76 +342,77 @@ class REKA(Provider):
                     f"Optimizer is not one of {self.__available_optimizers}"
                 )
 
-        self.session.headers.update(self.headers)
         payload = {
-
-            "conversation_history": [
-                {"type": "human", "text": f"## SYSTEM PROMPT: {self.system_prompt}\n\n## QUERY: {conversation_prompt}"},
-            ],
-
-            "stream": stream,
-            "use_search_engine": self.use_search_engine,
-            "use_code_interpreter": self.use_code_interpreter,
-            "model_name": self.model,
-            # "model_name": "reka-flash",
-            # "model_name": "reka-edge",
+            "prompt": f"{conversation_prompt}<s>[INST] {prompt} [/INST]",
+            "model": self.model,
+            "systemPrompt": "You are a helpful assistant.",
+            "temperature": self.temperature,
+            "topP": self.top_p,
+            "maxTokens": self.max_tokens_to_sample,
+            "image": None,
+            "audio": None,
         }
 
-        def for_stream():
-            response = self.session.post(self.api_endpoint, json=payload, stream=True, timeout=self.timeout)
-            if not response.ok:
-                raise Exception(
-                    f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
-                )
-
-            for value in response.iter_lines(
-                decode_unicode=True,
-                chunk_size=self.stream_chunk_size,
-            ):
-                try:
-                    resp = json.loads(value)
-                    self.last_response.update(resp)
-                    yield value if raw else resp
-                except json.decoder.JSONDecodeError:
-                    pass
+        async def for_stream():
+            async with self.session.stream(
+                "POST", self.chat_endpoint, json=payload, timeout=self.timeout
+            ) as response:
+                if (
+                    not response.is_success
+                    or not response.headers.get("Content-Type")
+                    == "text/plain; charset=utf-8"
+                ):
+                    raise exceptions.FailedToGenerateResponseError(
+                        f"Failed to generate response - ({response.status_code}, {response.reason_phrase})"
+                    )
+                message_load: str = ""
+                async for value in response.aiter_lines():
+                    try:
+                        if bool(value.strip()):
+                            message_load += value + "\n"
+                            resp: dict = dict(text=message_load)
+                            yield value if raw else resp
+                            self.last_response.update(resp)
+                    except json.decoder.JSONDecodeError:
+                        pass
             self.conversation.update_chat_history(
-                prompt, self.get_message(self.last_response)
+                prompt, await self.get_message(self.last_response)
             )
 
-        def for_non_stream():
-            # let's make use of stream
-            for _ in for_stream():
+        async def for_non_stream():
+            async for _ in for_stream():
                 pass
             return self.last_response
 
-        return for_stream() if stream else for_non_stream()
+        return for_stream() if stream else await for_non_stream()
 
-    def chat(
+    async def chat(
         self,
         prompt: str,
         stream: bool = False,
         optimizer: str = None,
         conversationally: bool = False,
-    ) -> str:
-        """Generate response `str`
+    ) -> str | AsyncGenerator:
+        """Generate response `str` asynchronously.
         Args:
             prompt (str): Prompt to be send.
             stream (bool, optional): Flag for streaming response. Defaults to False.
             optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
             conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
         Returns:
-            str: Response generated
+            str|AsyncGenerator: Response generated
         """
 
-        def for_stream():
-            for response in self.ask(
+        async def for_stream():
+            async_ask = await self.ask(
                 prompt, True, optimizer=optimizer, conversationally=conversationally
-            ):
-                yield self.get_message(response)
+            )
+            async for response in async_ask:
+                yield await self.get_message(response)
 
-        def for_non_stream():
-            return self.get_message(
-                self.ask(
+        async def for_non_stream():
+            return await self.get_message(
+                await self.ask(
                     prompt,
                     False,
                     optimizer=optimizer,
@@ -401,31 +420,29 @@ class REKA(Provider):
                 )
             )
 
-        return for_stream() if stream else for_non_stream()
+        return for_stream() if stream else await for_non_stream()
 
-    def get_message(self, response: dict) -> str:
+    async def get_message(self, response: dict) -> str:
         """Retrieves message only from response
 
         Args:
-            response (dict): Response generated by `self.ask`
+            response (str): Response generated by `self.ask`
 
         Returns:
             str: Message extracted
         """
         assert isinstance(response, dict), "Response should be of dict data-type only"
-        return response.get("text")
-#-----------------------------------------------GROQ-----------------------------------------------
-class GROQ(Provider):
+        return response["text"]
+#-----------------------------------------------Cohere--------------------------------------------
+class Cohere(Provider):
     def __init__(
         self,
         api_key: str,
         is_conversation: bool = True,
         max_tokens: int = 600,
-        temperature: float = 1,
-        presence_penalty: int = 0,
-        frequency_penalty: int = 0,
-        top_p: float = 1,
-        model: str = "mixtral-8x7b-32768",
+        model: str = "command-r-plus",
+        temperature: float = 0.7,
+        system_prompt: str = "You are helpful AI",
         timeout: int = 30,
         intro: str = None,
         filepath: str = None,
@@ -433,18 +450,20 @@ class GROQ(Provider):
         proxies: dict = {},
         history_offset: int = 10250,
         act: str = None,
+        top_k: int = -1,
+        top_p: float = 0.999,
     ):
-        """Instantiates GROQ
+        """Initializes Cohere
 
         Args:
-            api_key (key): GROQ's API key.
+            api_key (str): Cohere API key.
             is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True.
             max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 600.
-            temperature (float, optional): Charge of the generated text's randomness. Defaults to 1.
-            presence_penalty (int, optional): Chances of topic being repeated. Defaults to 0.
-            frequency_penalty (int, optional): Chances of word being repeated. Defaults to 0.
-            top_p (float, optional): Sampling threshold during inference time. Defaults to 0.999.
-            model (str, optional): LLM model name. Defaults to "gpt-3.5-turbo".
+            model (str, optional): Model to use for generating text. Defaults to "command-r-plus".
+            temperature (float, optional): Diversity of the generated text. Higher values produce more diverse outputs.
+                            Defaults to 0.7.
+            system_prompt (str, optional): A system_prompt or context to set the style or tone of the generated text. 
+                            Defaults to "You are helpful AI".
             timeout (int, optional): Http request timeout. Defaults to 30.
             intro (str, optional): Conversation introductory prompt. Defaults to None.
             filepath (str, optional): Path to file containing conversation history. Defaults to None.
@@ -459,10 +478,8 @@ class GROQ(Provider):
         self.api_key = api_key
         self.model = model
         self.temperature = temperature
-        self.presence_penalty = presence_penalty
-        self.frequency_penalty = frequency_penalty
-        self.top_p = top_p
-        self.chat_endpoint = "https://api.groq.com/openai/v1/chat/completions"
+        self.system_prompt = system_prompt
+        self.chat_endpoint = "https://production.api.os.cohere.ai/coral/v1/chat"
         self.stream_chunk_size = 64
         self.timeout = timeout
         self.last_response = {}
@@ -500,42 +517,19 @@ class GROQ(Provider):
     ) -> dict:
         """Chat with AI
 
-                Args:
-                    prompt (str): Prompt to be send.
-                    stream (bool, optional): Flag for streaming response. Defaults to False.
-                    raw (bool, optional): Stream back raw response as received. Defaults to False.
-                    optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
-                    conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
-                Returns:
-                   dict : {}
-                ```json
+        Args:
+            prompt (str): Prompt to be send.
+            stream (bool, optional): Flag for streaming response. Defaults to False.
+            raw (bool, optional): Stream back raw response as received. Defaults to False.
+            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+        Returns:
+           dict : {}
+        ```json
         {
-            "id": "c0c8d139-d2b9-9909-8aa1-14948bc28404",
-            "object": "chat.completion",
-            "created": 1710852779,
-            "model": "mixtral-8x7b-32768",
-            "choices": [
-                {
-                    "index": 0,
-                    "message": {
-                        "role": "assistant",
-                        "content": "Hello! How can I assist you today? I'm here to help answer your questions and engage in conversation on a wide variety of topics. Feel free to ask me anything!"
-                    },
-                    "logprobs": null,
-                    "finish_reason": "stop"
-                }
-            ],
-            "usage": {
-                "prompt_tokens": 47,
-                "prompt_time": 0.03,
-                "completion_tokens": 37,
-                "completion_time": 0.069,
-                "total_tokens": 84,
-                "total_time": 0.099
-            },
-            "system_fingerprint": null
+           "text" : "How may I assist you today?"
         }
-                ```
+        ```
         """
         conversation_prompt = self.conversation.gen_complete_prompt(prompt)
         if optimizer:
@@ -549,13 +543,10 @@ class GROQ(Provider):
                 )
         self.session.headers.update(self.headers)
         payload = {
-            "frequency_penalty": self.frequency_penalty,
-            "messages": [{"content": conversation_prompt, "role": "user"}],
+            "message": conversation_prompt,
             "model": self.model,
-            "presence_penalty": self.presence_penalty,
-            "stream": stream,
             "temperature": self.temperature,
-            "top_p": self.top_p,
+            "preamble": self.system_prompt,
         }
 
         def for_stream():
@@ -567,22 +558,14 @@ class GROQ(Provider):
                     f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
                 )
 
-            message_load = ""
             for value in response.iter_lines(
                 decode_unicode=True,
-                delimiter="" if raw else "data:",
                 chunk_size=self.stream_chunk_size,
             ):
                 try:
-                    resp = json.loads(value)
-                    incomplete_message = self.get_message(resp)
-                    if incomplete_message:
-                        message_load += incomplete_message
-                        resp["choices"][0]["delta"]["content"] = message_load
-                        self.last_response.update(resp)
-                        yield value if raw else resp
-                    elif raw:
-                        yield value
+                    resp = json.loads(value.strip().split("\n")[-1])
+                    self.last_response.update(resp)
+                    yield value if raw else resp
                 except json.decoder.JSONDecodeError:
                     pass
             self.conversation.update_chat_history(
@@ -590,19 +573,10 @@ class GROQ(Provider):
             )
 
         def for_non_stream():
-            response = self.session.post(
-                self.chat_endpoint, json=payload, stream=False, timeout=self.timeout
-            )
-            if not response.ok:
-                raise Exception(
-                    f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
-                )
-            resp = response.json()
-            self.last_response.update(resp)
-            self.conversation.update_chat_history(
-                prompt, self.get_message(self.last_response)
-            )
-            return resp
+            # let's make use of stream
+            for _ in for_stream():
+                pass
+            return self.last_response
 
         return for_stream() if stream else for_non_stream()
 
@@ -651,15 +625,12 @@ class GROQ(Provider):
             str: Message extracted
         """
         assert isinstance(response, dict), "Response should be of dict data-type only"
-        try:
-            if response["choices"][0].get("delta"):
-                return response["choices"][0]["delta"]["content"]
-            return response["choices"][0]["message"]["content"]
-        except KeyError:
-            return ""
-#----------------------------------------------------------Sean-------------------------------------class Sean:
+        return response["result"]["chatStreamEndEvent"]["response"]["text"]
+#-----------------------------------------------REKA-----------------------------------------------
+class REKA(Provider):
     def __init__(
         self,
+        api_key: str,
         is_conversation: bool = True,
         max_tokens: int = 600,
         timeout: int = 30,
@@ -669,8 +640,12 @@ class GROQ(Provider):
         proxies: dict = {},
         history_offset: int = 10250,
         act: str = None,
+        model: str = "reka-core",
+        system_prompt: str = "Be Helpful and Friendly. Keep your response straightforward, short and concise",
+        use_search_engine: bool = False,
+        use_code_interpreter: bool = False,
     ):
-        """Instantiates OPENGPT
+        """Instantiates REKA
 
         Args:
             is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True
@@ -682,31 +657,25 @@ class GROQ(Provider):
             proxies (dict, optional): Http request proxies. Defaults to {}.
             history_offset (int, optional): Limit conversation history to this number of last texts. Defaults to 10250.
             act (str|int, optional): Awesome prompt key or index. (Used as intro). Defaults to None.
+            model (str, optional): REKA model name. Defaults to "reka-core".
+            system_prompt (str, optional): System prompt for REKA. Defaults to "Be Helpful and Friendly. Keep your response straightforward, short and concise".
+            use_search_engine (bool, optional): Whether to use the search engine. Defaults to False.
+            use_code_interpreter (bool, optional): Whether to use the code interpreter. Defaults to False.
         """
         self.session = requests.Session()
-        self.max_tokens_to_sample = max_tokens
         self.is_conversation = is_conversation
-        self.chat_endpoint = (
-            "https://opengpts-example-vz4y4ooboq-uc.a.run.app/runs/stream"
-        )
+        self.max_tokens_to_sample = max_tokens
+        self.api_endpoint = "https://chat.reka.ai/api/chat"
         self.stream_chunk_size = 64
         self.timeout = timeout
         self.last_response = {}
-        self.assistant_id = "281bc620-b9f3-47c6-bf74-3f0e5b6e7dac"
-        self.authority = "opengpts-example-vz4y4ooboq-uc.a.run.app"
-
+        self.model = model
+        self.system_prompt = system_prompt
+        self.use_search_engine = use_search_engine
+        self.use_code_interpreter = use_code_interpreter
+        self.access_token = api_key
         self.headers = {
-            "authority": self.authority,
-            "accept": "text/event-stream",
-            "accept-language": "en-US,en;q=0.7",
-            "cache-control": "no-cache",
-            "content-type": "application/json",
-            "origin": "https://opengpts-example-vz4y4ooboq-uc.a.run.app",
-            "pragma": "no-cache",
-            "referer": "https://opengpts-example-vz4y4ooboq-uc.a.run.app/",
-            "sec-fetch-site": "same-origin",
-            "sec-gpc": "1",
-            "user-agent": "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36",
+            "Authorization": f"Bearer {self.access_token}",
         }
 
         self.__available_optimizers = (
@@ -748,27 +717,7 @@ class GROQ(Provider):
            dict : {}
         ```json
         {
-            "messages": [
-                {
-                    "content": "Hello there",
-                    "additional_kwargs": {},
-                    "type": "human",
-                    "example": false
-                },
-                {
-                    "content": "Hello! How can I assist you today?",
-                    "additional_kwargs": {
-                    "agent": {
-                        "return_values": {
-                            "output": "Hello! How can I assist you today?"
-                            },
-                        "log": "Hello! How can I assist you today?",
-                        "type": "AgentFinish"
-                    }
-                },
-                "type": "ai",
-                "example": false
-                }]
+           "text" : "How may I assist you today?"
         }
         ```
         """
@@ -784,33 +733,23 @@ class GROQ(Provider):
                 )
 
         self.session.headers.update(self.headers)
-        self.session.headers.update(
-            dict(
-                cookie=f"opengpts_user_id={uuid4().__str__()}",
-            )
-        )
         payload = {
-            "input": [
-                {
-                    "content": conversation_prompt,
-                    "additional_kwargs": {},
-                    "type": "human",
-                    "example": False,
-                },
+
+            "conversation_history": [
+                {"type": "human", "text": f"## SYSTEM PROMPT: {self.system_prompt}\n\n## QUERY: {conversation_prompt}"},
             ],
-            "assistant_id": self.assistant_id,
-            "thread_id": "",
+
+            "stream": stream,
+            "use_search_engine": self.use_search_engine,
+            "use_code_interpreter": self.use_code_interpreter,
+            "model_name": self.model,
+            # "model_name": "reka-flash",
+            # "model_name": "reka-edge",
         }
 
         def for_stream():
-            response = self.session.post(
-                self.chat_endpoint, json=payload, stream=True, timeout=self.timeout
-            )
-            if (
-                not response.ok
-                or not response.headers.get("Content-Type")
-                == "text/event-stream; charset=utf-8"
-            ):
+            response = self.session.post(self.api_endpoint, json=payload, stream=True, timeout=self.timeout)
+            if not response.ok:
                 raise Exception(
                     f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
                 )
@@ -820,12 +759,9 @@ class GROQ(Provider):
                 chunk_size=self.stream_chunk_size,
             ):
                 try:
-                    modified_value = re.sub("data:", "", value)
-                    resp = json.loads(modified_value)
-                    if len(resp) == 1:
-                        continue
-                    self.last_response.update(resp[1])
-                    yield value if raw else resp[1]
+                    resp = json.loads(value)
+                    self.last_response.update(resp)
+                    yield value if raw else resp
                 except json.decoder.JSONDecodeError:
                     pass
             self.conversation.update_chat_history(
@@ -833,6 +769,7 @@ class GROQ(Provider):
             )
 
         def for_non_stream():
+            # let's make use of stream
             for _ in for_stream():
                 pass
             return self.last_response
@@ -884,10 +821,9 @@ class GROQ(Provider):
             str: Message extracted
         """
         assert isinstance(response, dict), "Response should be of dict data-type only"
-        return response["content"]
-#----------------------------------------------------------OpenAI-----------------------------------
-class OPENAI(Provider):
-    model = "gpt-3.5-turbo"
+        return response.get("text")
+#-----------------------------------------------GROQ-----------------------------------------------
+class GROQ(Provider):
     def __init__(
         self,
         api_key: str,
@@ -897,7 +833,7 @@ class OPENAI(Provider):
         presence_penalty: int = 0,
         frequency_penalty: int = 0,
         top_p: float = 1,
-        model: str = model,
+        model: str = "mixtral-8x7b-32768",
         timeout: int = 30,
         intro: str = None,
         filepath: str = None,
@@ -906,10 +842,10 @@ class OPENAI(Provider):
         history_offset: int = 10250,
         act: str = None,
     ):
-        """Instantiates OPENAI
+        """Instantiates GROQ
 
         Args:
-            api_key (key): OpenAI's API key.
+            api_key (key): GROQ's API key.
             is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True.
             max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 600.
             temperature (float, optional): Charge of the generated text's randomness. Defaults to 1.
@@ -934,7 +870,7 @@ class OPENAI(Provider):
         self.presence_penalty = presence_penalty
         self.frequency_penalty = frequency_penalty
         self.top_p = top_p
-        self.chat_endpoint = "https://api.openai.com/v1/chat/completions"
+        self.chat_endpoint = "https://api.groq.com/openai/v1/chat/completions"
         self.stream_chunk_size = 64
         self.timeout = timeout
         self.last_response = {}
@@ -972,37 +908,42 @@ class OPENAI(Provider):
     ) -> dict:
         """Chat with AI
 
-        Args:
-            prompt (str): Prompt to be send.
-            stream (bool, optional): Flag for streaming response. Defaults to False.
-            raw (bool, optional): Stream back raw response as received. Defaults to False.
-            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
-            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
-        Returns:
-           dict : {}
-        ```json
+                Args:
+                    prompt (str): Prompt to be send.
+                    stream (bool, optional): Flag for streaming response. Defaults to False.
+                    raw (bool, optional): Stream back raw response as received. Defaults to False.
+                    optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+                    conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+                Returns:
+                   dict : {}
+                ```json
         {
-            "id": "chatcmpl-TaREJpBZsRVQFRFic1wIA7Q7XfnaD",
+            "id": "c0c8d139-d2b9-9909-8aa1-14948bc28404",
             "object": "chat.completion",
-            "created": 1704623244,
-            "model": "gpt-3.5-turbo",
-            "usage": {
-                "prompt_tokens": 0,
-                "completion_tokens": 0,
-                "total_tokens": 0
-                },
+            "created": 1710852779,
+            "model": "mixtral-8x7b-32768",
             "choices": [
                 {
+                    "index": 0,
                     "message": {
                         "role": "assistant",
-                        "content": "Hello! How can I assist you today?"
-                },
-                "finish_reason": "stop",
-                "index": 0
+                        "content": "Hello! How can I assist you today? I'm here to help answer your questions and engage in conversation on a wide variety of topics. Feel free to ask me anything!"
+                    },
+                    "logprobs": null,
+                    "finish_reason": "stop"
                 }
-            ]
+            ],
+            "usage": {
+                "prompt_tokens": 47,
+                "prompt_time": 0.03,
+                "completion_tokens": 37,
+                "completion_time": 0.069,
+                "total_tokens": 84,
+                "total_time": 0.099
+            },
+            "system_fingerprint": null
         }
-        ```
+                ```
         """
         conversation_prompt = self.conversation.gen_complete_prompt(prompt)
         if optimizer:
@@ -1060,10 +1001,7 @@ class OPENAI(Provider):
             response = self.session.post(
                 self.chat_endpoint, json=payload, stream=False, timeout=self.timeout
             )
-            if (
-                not response.ok
-                or not response.headers.get("Content-Type", "") == "application/json"
-            ):
+            if not response.ok:
                 raise Exception(
                     f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
                 )
@@ -1127,21 +1065,17 @@ class OPENAI(Provider):
             return response["choices"][0]["message"]["content"]
         except KeyError:
             return ""
-#--------------------------------------LEO-----------------------------------------
-class LEO(Provider):
-    
-    # model = "llama-2-13b-chat"
-
-    # key = "qztbjzBqJueQZLFkwTTJrieu8Vw3789u"
+class AsyncGROQ(AsyncProvider):
     def __init__(
         self,
+        api_key: str,
         is_conversation: bool = True,
         max_tokens: int = 600,
-        temperature: float = 0.2,
-        top_k: int = -1,
-        top_p: float = 0.999,
-        model: str = "llama-2-13b-chat",
-        brave_key: str = "qztbjzBqJueQZLFkwTTJrieu8Vw3789u",
+        temperature: float = 1,
+        presence_penalty: int = 0,
+        frequency_penalty: int = 0,
+        top_p: float = 1,
+        model: str = "mixtral-8x7b-32768",
         timeout: int = 30,
         intro: str = None,
         filepath: str = None,
@@ -1150,49 +1084,47 @@ class LEO(Provider):
         history_offset: int = 10250,
         act: str = None,
     ):
-        """Instantiate TGPT
+        """Instantiates GROQ
 
         Args:
-            is_conversation (str, optional): Flag for chatting conversationally. Defaults to True.
-            brave_key (str, optional): Brave API access key. Defaults to "qztbjzBqJueQZLFkwTTJrieu8Vw3789u".
-            model (str, optional): Text generation model name. Defaults to "llama-2-13b-chat".
+            api_key (key): GROQ's API key.
+            is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True.
             max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 600.
-            temperature (float, optional): Charge of the generated text's randomness. Defaults to 0.2.
-            top_k (int, optional): Chance of topic being repeated. Defaults to -1.
+            temperature (float, optional): Charge of the generated text's randomness. Defaults to 1.
+            presence_penalty (int, optional): Chances of topic being repeated. Defaults to 0.
+            frequency_penalty (int, optional): Chances of word being repeated. Defaults to 0.
             top_p (float, optional): Sampling threshold during inference time. Defaults to 0.999.
-            timeput (int, optional): Http requesting timeout. Defaults to 30
-            intro (str, optional): Conversation introductory prompt. Defaults to `Conversation.intro`.
+            model (str, optional): LLM model name. Defaults to "gpt-3.5-turbo".
+            timeout (int, optional): Http request timeout. Defaults to 30.
+            intro (str, optional): Conversation introductory prompt. Defaults to None.
             filepath (str, optional): Path to file containing conversation history. Defaults to None.
             update_file (bool, optional): Add new prompts and responses to the file. Defaults to True.
-            proxies (dict, optional) : Http reqiuest proxies (socks). Defaults to {}.
+            proxies (dict, optional): Http request proxies. Defaults to {}.
             history_offset (int, optional): Limit conversation history to this number of last texts. Defaults to 10250.
             act (str|int, optional): Awesome prompt key or index. (Used as intro). Defaults to None.
         """
-        self.session = requests.Session()
         self.is_conversation = is_conversation
         self.max_tokens_to_sample = max_tokens
+        self.api_key = api_key
         self.model = model
-        self.stop_sequences = ["</response>", "</s>"]
         self.temperature = temperature
-        self.top_k = top_k
+        self.presence_penalty = presence_penalty
+        self.frequency_penalty = frequency_penalty
         self.top_p = top_p
-        self.chat_endpoint = "https://ai-chat.bsg.brave.com/v1/complete"
+        self.chat_endpoint = "https://api.groq.com/openai/v1/chat/completions"
         self.stream_chunk_size = 64
         self.timeout = timeout
         self.last_response = {}
         self.headers = {
             "Content-Type": "application/json",
-            "accept": "text/event-stream",
-            "x-brave-key": brave_key,
-            "accept-language": "en-US,en;q=0.9",
-            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:99.0) Gecko/20100101 Firefox/110.0",
+            "Authorization": f"Bearer {self.api_key}",
         }
+
         self.__available_optimizers = (
             method
             for method in dir(Optimizers)
             if callable(getattr(Optimizers, method)) and not method.startswith("__")
         )
-        self.session.headers.update(self.headers)
         Conversation.intro = (
             AwesomePrompts().get_act(
                 act, raise_not_found=True, default=None, case_insensitive=True
@@ -1204,45 +1136,54 @@ class LEO(Provider):
             is_conversation, self.max_tokens_to_sample, filepath, update_file
         )
         self.conversation.history_offset = history_offset
-        self.session.proxies = proxies
-        self.system_prompt = (
-            "\n\nYour name is Leo, a helpful"
-            "respectful and honest AI assistant created by the company Brave. You will be replying to a user of the Brave browser. "
-            "Always respond in a neutral tone. Be polite and courteous. Answer concisely in no more than 50-80 words."
-            "\n\nPlease ensure that your responses are socially unbiased and positive in nature."
-            "If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. "
-            "If you don't know the answer to a question, please don't share false information.\n"
-        )
+        self.session = httpx.AsyncClient(headers=self.headers, proxies=proxies)
 
-    def ask(
+    async def ask(
         self,
         prompt: str,
         stream: bool = False,
         raw: bool = False,
         optimizer: str = None,
         conversationally: bool = False,
-    ) -> dict:
-        """Chat with AI
+    ) -> dict | AsyncGenerator:
+        """Chat with AI asynchronously.
 
-        Args:
-            prompt (str): Prompt to be send.
-            stream (bool, optional): Flag for streaming response. Defaults to False.
-            raw (bool, optional): Stream back raw response as received. Defaults to False.
-            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
-            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
-        Returns:
-           dict : {}
-        ```json
+                Args:
+                    prompt (str): Prompt to be send.
+                    stream (bool, optional): Flag for streaming response. Defaults to False.
+                    raw (bool, optional): Stream back raw response as received. Defaults to False.
+                    optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+                    conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+                Returns:
+                   dict|AsyncGenerator : ai content
+                ```json
         {
-            "completion": "\nNext: domestic cat breeds with short hair >>",
-            "stop_reason": null,
-            "truncated": false,
-            "stop": null,
-            "model": "llama-2-13b-chat",
-            "log_id": "cmpl-3kYiYxSNDvgMShSzFooz6t",
-            "exception": null
+            "id": "c0c8d139-d2b9-9909-8aa1-14948bc28404",
+            "object": "chat.completion",
+            "created": 1710852779,
+            "model": "mixtral-8x7b-32768",
+            "choices": [
+                {
+                    "index": 0,
+                    "message": {
+                        "role": "assistant",
+                        "content": "Hello! How can I assist you today? I'm here to help answer your questions and engage in conversation on a wide variety of topics. Feel free to ask me anything!"
+                    },
+                    "logprobs": null,
+                    "finish_reason": "stop"
+                }
+            ],
+            "usage": {
+                "prompt_tokens": 47,
+                "prompt_time": 0.03,
+                "completion_tokens": 37,
+                "completion_time": 0.069,
+                "total_tokens": 84,
+                "total_time": 0.099
+            },
+            "system_fingerprint": null
         }
-        ```
+                ```
         """
         conversation_prompt = self.conversation.gen_complete_prompt(prompt)
         if optimizer:
@@ -1254,92 +1195,90 @@ class LEO(Provider):
                 raise Exception(
                     f"Optimizer is not one of {self.__available_optimizers}"
                 )
-
-        self.session.headers.update(self.headers)
         payload = {
-            "max_tokens_to_sample": self.max_tokens_to_sample,
+            "frequency_penalty": self.frequency_penalty,
+            "messages": [{"content": conversation_prompt, "role": "user"}],
             "model": self.model,
-            "prompt": f"<s>[INST] <<SYS>>{self.system_prompt}<</SYS>>{conversation_prompt} [/INST]",
-            "self.stop_sequence": self.stop_sequences,
+            "presence_penalty": self.presence_penalty,
             "stream": stream,
-            "top_k": self.top_k,
+            "temperature": self.temperature,
             "top_p": self.top_p,
         }
 
-        def for_stream():
-            response = self.session.post(
-                self.chat_endpoint, json=payload, stream=True, timeout=self.timeout
-            )
-            if (
-                not response.ok
-                or not response.headers.get("Content-Type")
-                == "text/event-stream; charset=utf-8"
-            ):
-                raise Exception(
-                    f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
-                )
-
-            for value in response.iter_lines(
-                decode_unicode=True,
-                delimiter="" if raw else "data:",
-                chunk_size=self.stream_chunk_size,
-            ):
-                try:
-                    resp = json.loads(value)
-                    self.last_response.update(resp)
-                    yield value if raw else resp
-                except json.decoder.JSONDecodeError:
-                    pass
+        async def for_stream():
+            async with self.session.stream(
+                "POST", self.chat_endpoint, json=payload, timeout=self.timeout
+            ) as response:
+                if not response.is_success:
+                    raise Exception(
+                        f"Failed to generate response - ({response.status_code}, {response.reason_phrase})"
+                    )
+
+                message_load = ""
+                intro_value = "data:"
+                async for value in response.aiter_lines():
+                    try:
+                        if value.startswith(intro_value):
+                            value = value[len(intro_value) :]
+                        resp = json.loads(value)
+                        incomplete_message = await self.get_message(resp)
+                        if incomplete_message:
+                            message_load += incomplete_message
+                            resp["choices"][0]["delta"]["content"] = message_load
+                            self.last_response.update(resp)
+                            yield value if raw else resp
+                        elif raw:
+                            yield value
+                    except json.decoder.JSONDecodeError:
+                        pass
             self.conversation.update_chat_history(
-                prompt, self.get_message(self.last_response)
+                prompt, await self.get_message(self.last_response)
             )
 
-        def for_non_stream():
-            response = self.session.post(
-                self.chat_endpoint, json=payload, stream=False, timeout=self.timeout
+        async def for_non_stream():
+            response = httpx.post(
+                self.chat_endpoint, json=payload, timeout=self.timeout
             )
-            if (
-                not response.ok
-                or not response.headers.get("Content-Type", "") == "application/json"
-            ):
+            if not response.is_success:
                 raise Exception(
-                    f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
+                    f"Failed to generate response - ({response.status_code}, {response.reason_phrase})"
                 )
             resp = response.json()
             self.last_response.update(resp)
             self.conversation.update_chat_history(
-                prompt, self.get_message(self.last_response)
+                prompt, await self.get_message(self.last_response)
             )
             return resp
 
-        return for_stream() if stream else for_non_stream()
+        return for_stream() if stream else await for_non_stream()
 
-    def chat(
+    async def chat(
         self,
         prompt: str,
         stream: bool = False,
         optimizer: str = None,
         conversationally: bool = False,
-    ) -> str:
-        """Generate response `str`
+    ) -> str | AsyncGenerator:
+        """Generate response `str` asynchronously.
         Args:
             prompt (str): Prompt to be send.
             stream (bool, optional): Flag for streaming response. Defaults to False.
             optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
             conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
         Returns:
-            str: Response generated
+            str|AsyncGenerator: Response generated
         """
 
-        def for_stream():
-            for response in self.ask(
+        async def for_stream():
+            async_ask = await self.ask(
                 prompt, True, optimizer=optimizer, conversationally=conversationally
-            ):
-                yield self.get_message(response)
+            )
+            async for response in async_ask:
+                yield await self.get_message(response)
 
-        def for_non_stream():
-            return self.get_message(
-                self.ask(
+        async def for_non_stream():
+            return await self.get_message(
+                await self.ask(
                     prompt,
                     False,
                     optimizer=optimizer,
@@ -1347,9 +1286,9 @@ class LEO(Provider):
                 )
             )
 
-        return for_stream() if stream else for_non_stream()
+        return for_stream() if stream else await for_non_stream()
 
-    def get_message(self, response: dict) -> str:
+    async def get_message(self, response: dict) -> str:
         """Retrieves message only from response
 
         Args:
@@ -1359,15 +1298,24 @@ class LEO(Provider):
             str: Message extracted
         """
         assert isinstance(response, dict), "Response should be of dict data-type only"
-        return response.get("completion")
-#------------------------------------------------------KOBOLDAI-----------------------------------------------------------
-class KOBOLDAI(Provider):
+        try:
+            if response["choices"][0].get("delta"):
+                return response["choices"][0]["delta"]["content"]
+            return response["choices"][0]["message"]["content"]
+        except KeyError:
+            return ""
+#----------------------------------------------------------OpenAI-----------------------------------
+class OPENAI(Provider):
     def __init__(
         self,
+        api_key: str,
         is_conversation: bool = True,
         max_tokens: int = 600,
         temperature: float = 1,
+        presence_penalty: int = 0,
+        frequency_penalty: int = 0,
         top_p: float = 1,
+        model: str = "gpt-3.5-turbo",
         timeout: int = 30,
         intro: str = None,
         filepath: str = None,
@@ -1376,35 +1324,40 @@ class KOBOLDAI(Provider):
         history_offset: int = 10250,
         act: str = None,
     ):
-        """Instantiate TGPT
+        """Instantiates OPENAI
 
         Args:
-            is_conversation (str, optional): Flag for chatting conversationally. Defaults to True.
+            api_key (key): OpenAI's API key.
+            is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True.
             max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 600.
-            temperature (float, optional): Charge of the generated text's randomness. Defaults to 0.2.
+            temperature (float, optional): Charge of the generated text's randomness. Defaults to 1.
+            presence_penalty (int, optional): Chances of topic being repeated. Defaults to 0.
+            frequency_penalty (int, optional): Chances of word being repeated. Defaults to 0.
             top_p (float, optional): Sampling threshold during inference time. Defaults to 0.999.
-            timeout (int, optional): Http requesting timeout. Defaults to 30
-            intro (str, optional): Conversation introductory prompt. Defaults to `Conversation.intro`.
+            model (str, optional): LLM model name. Defaults to "gpt-3.5-turbo".
+            timeout (int, optional): Http request timeout. Defaults to 30.
+            intro (str, optional): Conversation introductory prompt. Defaults to None.
             filepath (str, optional): Path to file containing conversation history. Defaults to None.
             update_file (bool, optional): Add new prompts and responses to the file. Defaults to True.
-            proxies (dict, optional) : Http reqiuest proxies (socks). Defaults to {}.
+            proxies (dict, optional): Http request proxies. Defaults to {}.
             history_offset (int, optional): Limit conversation history to this number of last texts. Defaults to 10250.
             act (str|int, optional): Awesome prompt key or index. (Used as intro). Defaults to None.
         """
-        self.session = requests.Session()
         self.is_conversation = is_conversation
         self.max_tokens_to_sample = max_tokens
+        self.api_key = api_key
+        self.model = model
         self.temperature = temperature
+        self.presence_penalty = presence_penalty
+        self.frequency_penalty = frequency_penalty
         self.top_p = top_p
-        self.chat_endpoint = (
-            "https://koboldai-koboldcpp-tiefighter.hf.space/api/extra/generate/stream"
-        )
+        self.chat_endpoint = "https://api.openai.com/v1/chat/completions"
         self.stream_chunk_size = 64
         self.timeout = timeout
         self.last_response = {}
         self.headers = {
             "Content-Type": "application/json",
-            "Accept": "application/json",
+            "Authorization": f"Bearer {self.api_key}",
         }
 
         self.__available_optimizers = (
@@ -1446,7 +1399,25 @@ class KOBOLDAI(Provider):
            dict : {}
         ```json
         {
-           "token" : "How may I assist you today?"
+            "id": "chatcmpl-TaREJpBZsRVQFRFic1wIA7Q7XfnaD",
+            "object": "chat.completion",
+            "created": 1704623244,
+            "model": "gpt-3.5-turbo",
+            "usage": {
+                "prompt_tokens": 0,
+                "completion_tokens": 0,
+                "total_tokens": 0
+                },
+            "choices": [
+                {
+                    "message": {
+                        "role": "assistant",
+                        "content": "Hello! How can I assist you today?"
+                },
+                "finish_reason": "stop",
+                "index": 0
+                }
+            ]
         }
         ```
         """
@@ -1457,13 +1428,16 @@ class KOBOLDAI(Provider):
                     conversation_prompt if conversationally else prompt
                 )
             else:
-                raise Exception(
+                raise exceptions.FailedToGenerateResponseError(
                     f"Optimizer is not one of {self.__available_optimizers}"
                 )
-
         self.session.headers.update(self.headers)
         payload = {
-            "prompt": conversation_prompt,
+            "frequency_penalty": self.frequency_penalty,
+            "messages": [{"content": conversation_prompt, "role": "user"}],
+            "model": self.model,
+            "presence_penalty": self.presence_penalty,
+            "stream": stream,
             "temperature": self.temperature,
             "top_p": self.top_p,
         }
@@ -1473,22 +1447,26 @@ class KOBOLDAI(Provider):
                 self.chat_endpoint, json=payload, stream=True, timeout=self.timeout
             )
             if not response.ok:
-                raise Exception(
+                raise exceptions.FailedToGenerateResponseError(
                     f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
                 )
 
             message_load = ""
             for value in response.iter_lines(
                 decode_unicode=True,
-                delimiter="" if raw else "event: message\ndata:",
+                delimiter="" if raw else "data:",
                 chunk_size=self.stream_chunk_size,
             ):
                 try:
                     resp = json.loads(value)
-                    message_load += self.get_message(resp)
-                    resp["token"] = message_load
-                    self.last_response.update(resp)
-                    yield value if raw else resp
+                    incomplete_message = self.get_message(resp)
+                    if incomplete_message:
+                        message_load += incomplete_message
+                        resp["choices"][0]["delta"]["content"] = message_load
+                        self.last_response.update(resp)
+                        yield value if raw else resp
+                    elif raw:
+                        yield value
                 except json.decoder.JSONDecodeError:
                     pass
             self.conversation.update_chat_history(
@@ -1496,10 +1474,22 @@ class KOBOLDAI(Provider):
             )
 
         def for_non_stream():
-            # let's make use of stream
-            for _ in for_stream():
-                pass
-            return self.last_response
+            response = self.session.post(
+                self.chat_endpoint, json=payload, stream=False, timeout=self.timeout
+            )
+            if (
+                not response.ok
+                or not response.headers.get("Content-Type", "") == "application/json"
+            ):
+                raise exceptions.FailedToGenerateResponseError(
+                    f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
+                )
+            resp = response.json()
+            self.last_response.update(resp)
+            self.conversation.update_chat_history(
+                prompt, self.get_message(self.last_response)
+            )
+            return resp
 
         return for_stream() if stream else for_non_stream()
 
@@ -1548,13 +1538,23 @@ class KOBOLDAI(Provider):
             str: Message extracted
         """
         assert isinstance(response, dict), "Response should be of dict data-type only"
-        return response.get("token")
-#------------------------------------------------------OpenGPT-----------------------------------------------------------
-class OPENGPT:
+        try:
+            if response["choices"][0].get("delta"):
+                return response["choices"][0]["delta"]["content"]
+            return response["choices"][0]["message"]["content"]
+        except KeyError:
+            return ""
+class AsyncOPENAI(AsyncProvider):
     def __init__(
         self,
+        api_key: str,
         is_conversation: bool = True,
         max_tokens: int = 600,
+        temperature: float = 1,
+        presence_penalty: int = 0,
+        frequency_penalty: int = 0,
+        top_p: float = 1,
+        model: str = "gpt-3.5-turbo",
         timeout: int = 30,
         intro: str = None,
         filepath: str = None,
@@ -1563,11 +1563,17 @@ class OPENGPT:
         history_offset: int = 10250,
         act: str = None,
     ):
-        """Instantiates OPENGPT
+        """Instantiates OPENAI
 
         Args:
-            is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True
+            api_key (key): OpenAI's API key.
+            is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True.
             max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 600.
+            temperature (float, optional): Charge of the generated text's randomness. Defaults to 1.
+            presence_penalty (int, optional): Chances of topic being repeated. Defaults to 0.
+            frequency_penalty (int, optional): Chances of word being repeated. Defaults to 0.
+            top_p (float, optional): Sampling threshold during inference time. Defaults to 0.999.
+            model (str, optional): LLM model name. Defaults to "gpt-3.5-turbo".
             timeout (int, optional): Http request timeout. Defaults to 30.
             intro (str, optional): Conversation introductory prompt. Defaults to None.
             filepath (str, optional): Path to file containing conversation history. Defaults to None.
@@ -1576,31 +1582,2133 @@ class OPENGPT:
             history_offset (int, optional): Limit conversation history to this number of last texts. Defaults to 10250.
             act (str|int, optional): Awesome prompt key or index. (Used as intro). Defaults to None.
         """
-        self.session = requests.Session()
-        self.max_tokens_to_sample = max_tokens
         self.is_conversation = is_conversation
-        self.chat_endpoint = (
-            "https://opengpts-example-vz4y4ooboq-uc.a.run.app/runs/stream"
-        )
+        self.max_tokens_to_sample = max_tokens
+        self.api_key = api_key
+        self.model = model
+        self.temperature = temperature
+        self.presence_penalty = presence_penalty
+        self.frequency_penalty = frequency_penalty
+        self.top_p = top_p
+        self.chat_endpoint = "https://api.openai.com/v1/chat/completions"
         self.stream_chunk_size = 64
         self.timeout = timeout
         self.last_response = {}
-        self.assistant_id = "bca37014-6f97-4f2b-8928-81ea8d478d88"
-        self.authority = "opengpts-example-vz4y4ooboq-uc.a.run.app"
-
         self.headers = {
-            "authority": self.authority,
-            "accept": "text/event-stream",
-            "accept-language": "en-US,en;q=0.7",
-            "cache-control": "no-cache",
-            "content-type": "application/json",
-            "origin": "https://opengpts-example-vz4y4ooboq-uc.a.run.app",
-            "pragma": "no-cache",
-            "referer": "https://opengpts-example-vz4y4ooboq-uc.a.run.app/",
-            "sec-fetch-site": "same-origin",
-            "sec-gpc": "1",
-            "user-agent": "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36",
-        }
+            "Content-Type": "application/json",
+            "Authorization": f"Bearer {self.api_key}",
+        }
+
+        self.__available_optimizers = (
+            method
+            for method in dir(Optimizers)
+            if callable(getattr(Optimizers, method)) and not method.startswith("__")
+        )
+        Conversation.intro = (
+            AwesomePrompts().get_act(
+                act, raise_not_found=True, default=None, case_insensitive=True
+            )
+            if act
+            else intro or Conversation.intro
+        )
+        self.conversation = Conversation(
+            is_conversation, self.max_tokens_to_sample, filepath, update_file
+        )
+        self.conversation.history_offset = history_offset
+        self.session = httpx.AsyncClient(
+            headers=self.headers,
+            proxies=proxies,
+        )
+
+    async def ask(
+        self,
+        prompt: str,
+        stream: bool = False,
+        raw: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> dict | AsyncGenerator:
+        """Chat with AI asynchronously.
+
+        Args:
+            prompt (str): Prompt to be send.
+            stream (bool, optional): Flag for streaming response. Defaults to False.
+            raw (bool, optional): Stream back raw response as received. Defaults to False.
+            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+        Returns:
+           dict|AsyncGenerator : ai content.
+        ```json
+        {
+            "id": "chatcmpl-TaREJpBZsRVQFRFic1wIA7Q7XfnaD",
+            "object": "chat.completion",
+            "created": 1704623244,
+            "model": "gpt-3.5-turbo",
+            "usage": {
+                "prompt_tokens": 0,
+                "completion_tokens": 0,
+                "total_tokens": 0
+                },
+            "choices": [
+                {
+                    "message": {
+                        "role": "assistant",
+                        "content": "Hello! How can I assist you today?"
+                },
+                "finish_reason": "stop",
+                "index": 0
+                }
+            ]
+        }
+        ```
+        """
+        conversation_prompt = self.conversation.gen_complete_prompt(prompt)
+        if optimizer:
+            if optimizer in self.__available_optimizers:
+                conversation_prompt = getattr(Optimizers, optimizer)(
+                    conversation_prompt if conversationally else prompt
+                )
+            else:
+                raise Exception(
+                    f"Optimizer is not one of {self.__available_optimizers}"
+                )
+        payload = {
+            "frequency_penalty": self.frequency_penalty,
+            "messages": [{"content": conversation_prompt, "role": "user"}],
+            "model": self.model,
+            "presence_penalty": self.presence_penalty,
+            "stream": stream,
+            "temperature": self.temperature,
+            "top_p": self.top_p,
+        }
+
+        async def for_stream():
+            async with self.session.stream(
+                "POST", self.chat_endpoint, json=payload, timeout=self.timeout
+            ) as response:
+                if not response.is_success:
+                    raise Exception(
+                        f"Failed to generate response - ({response.status_code}, {response.reason_phrase})"
+                    )
+
+                message_load = ""
+                async for value in response.aiter_lines():
+                    try:
+
+                        resp = sanitize_stream(value)
+                        incomplete_message = await self.get_message(resp)
+                        if incomplete_message:
+                            message_load += incomplete_message
+                            resp["choices"][0]["delta"]["content"] = message_load
+                            self.last_response.update(resp)
+                            yield value if raw else resp
+                        elif raw:
+                            yield value
+                    except json.decoder.JSONDecodeError:
+                        pass
+            self.conversation.update_chat_history(
+                prompt, await self.get_message(self.last_response)
+            )
+
+        async def for_non_stream():
+            response = httpx.post(
+                self.chat_endpoint,
+                json=payload,
+                timeout=self.timeout,
+                headers=self.headers,
+            )
+            if (
+                not response.is_success
+                or not response.headers.get("Content-Type", "") == "application/json"
+            ):
+                raise Exception(
+                    f"Failed to generate response - ({response.status_code}, {response.reason_phrase})"
+                )
+            resp = response.json()
+            self.last_response.update(resp)
+            self.conversation.update_chat_history(
+                prompt, await self.get_message(self.last_response)
+            )
+            return resp
+
+        return for_stream() if stream else await for_non_stream()
+
+    async def chat(
+        self,
+        prompt: str,
+        stream: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> str | AsyncGenerator:
+        """Generate response `str` asynchronously.
+        Args:
+            prompt (str): Prompt to be send.
+            stream (bool, optional): Flag for streaming response. Defaults to False.
+            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+        Returns:
+            str|AsyncGenerator: Response generated
+        """
+
+        async def for_stream():
+            async_ask = await self.ask(
+                prompt, True, optimizer=optimizer, conversationally=conversationally
+            )
+            async for response in async_ask:
+                yield await self.get_message(response)
+
+        async def for_non_stream():
+            return await self.get_message(
+                await self.ask(
+                    prompt,
+                    False,
+                    optimizer=optimizer,
+                    conversationally=conversationally,
+                )
+            )
+
+        return for_stream() if stream else await for_non_stream()
+
+    async def get_message(self, response: dict) -> str:
+        """Retrieves message only from response asynchronously.
+
+        Args:
+            response (dict): Response generated by `self.ask`
+
+        Returns:
+            str: Message extracted
+        """
+        assert isinstance(response, dict), "Response should be of dict data-type only"
+        try:
+            if response["choices"][0].get("delta"):
+                return response["choices"][0]["delta"]["content"]
+            return response["choices"][0]["message"]["content"]
+        except KeyError:
+            return ""
+#--------------------------------------LEO-----------------------------------------
+class LEO(Provider):
+    
+    def __init__(
+        self,
+        is_conversation: bool = True,
+        max_tokens: int = 600,
+        temperature: float = 0.2,
+        top_k: int = -1,
+        top_p: float = 0.999,
+        model: str = "llama-2-13b-chat",
+        brave_key: str = "qztbjzBqJueQZLFkwTTJrieu8Vw3789u",
+        timeout: int = 30,
+        intro: str = None,
+        filepath: str = None,
+        update_file: bool = True,
+        proxies: dict = {},
+        history_offset: int = 10250,
+        act: str = None,
+    ):
+        """Instantiate TGPT
+
+        Args:
+            is_conversation (str, optional): Flag for chatting conversationally. Defaults to True.
+            brave_key (str, optional): Brave API access key. Defaults to "qztbjzBqJueQZLFkwTTJrieu8Vw3789u".
+            model (str, optional): Text generation model name. Defaults to "llama-2-13b-chat".
+            max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 600.
+            temperature (float, optional): Charge of the generated text's randomness. Defaults to 0.2.
+            top_k (int, optional): Chance of topic being repeated. Defaults to -1.
+            top_p (float, optional): Sampling threshold during inference time. Defaults to 0.999.
+            timeput (int, optional): Http requesting timeout. Defaults to 30
+            intro (str, optional): Conversation introductory prompt. Defaults to `Conversation.intro`.
+            filepath (str, optional): Path to file containing conversation history. Defaults to None.
+            update_file (bool, optional): Add new prompts and responses to the file. Defaults to True.
+            proxies (dict, optional) : Http reqiuest proxies (socks). Defaults to {}.
+            history_offset (int, optional): Limit conversation history to this number of last texts. Defaults to 10250.
+            act (str|int, optional): Awesome prompt key or index. (Used as intro). Defaults to None.
+        """
+        self.session = requests.Session()
+        self.is_conversation = is_conversation
+        self.max_tokens_to_sample = max_tokens
+        self.model = model
+        self.stop_sequences = ["</response>", "</s>"]
+        self.temperature = temperature
+        self.top_k = top_k
+        self.top_p = top_p
+        self.chat_endpoint = "https://ai-chat.bsg.brave.com/v1/complete"
+        self.stream_chunk_size = 64
+        self.timeout = timeout
+        self.last_response = {}
+        self.headers = {
+            "Content-Type": "application/json",
+            "accept": "text/event-stream",
+            "x-brave-key": brave_key,
+            "accept-language": "en-US,en;q=0.9",
+            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:99.0) Gecko/20100101 Firefox/110.0",
+        }
+        self.__available_optimizers = (
+            method
+            for method in dir(Optimizers)
+            if callable(getattr(Optimizers, method)) and not method.startswith("__")
+        )
+        self.session.headers.update(self.headers)
+        Conversation.intro = (
+            AwesomePrompts().get_act(
+                act, raise_not_found=True, default=None, case_insensitive=True
+            )
+            if act
+            else intro or Conversation.intro
+        )
+        self.conversation = Conversation(
+            is_conversation, self.max_tokens_to_sample, filepath, update_file
+        )
+        self.conversation.history_offset = history_offset
+        self.session.proxies = proxies
+        self.system_prompt = (
+            "\n\nYour name is Leo, a helpful"
+            "respectful and honest AI assistant created by the company Brave. You will be replying to a user of the Brave browser. "
+            "Always respond in a neutral tone. Be polite and courteous. Answer concisely in no more than 50-80 words."
+            "\n\nPlease ensure that your responses are socially unbiased and positive in nature."
+            "If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. "
+            "If you don't know the answer to a question, please don't share false information.\n"
+        )
+
+    def ask(
+        self,
+        prompt: str,
+        stream: bool = False,
+        raw: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> dict:
+        """Chat with AI
+
+        Args:
+            prompt (str): Prompt to be send.
+            stream (bool, optional): Flag for streaming response. Defaults to False.
+            raw (bool, optional): Stream back raw response as received. Defaults to False.
+            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+        Returns:
+           dict : {}
+        ```json
+        {
+            "completion": "\nNext: domestic cat breeds with short hair >>",
+            "stop_reason": null,
+            "truncated": false,
+            "stop": null,
+            "model": "llama-2-13b-chat",
+            "log_id": "cmpl-3kYiYxSNDvgMShSzFooz6t",
+            "exception": null
+        }
+        ```
+        """
+        conversation_prompt = self.conversation.gen_complete_prompt(prompt)
+        if optimizer:
+            if optimizer in self.__available_optimizers:
+                conversation_prompt = getattr(Optimizers, optimizer)(
+                    conversation_prompt if conversationally else prompt
+                )
+            else:
+                raise Exception(
+                    f"Optimizer is not one of {self.__available_optimizers}"
+                )
+
+        self.session.headers.update(self.headers)
+        payload = {
+            "max_tokens_to_sample": self.max_tokens_to_sample,
+            "model": self.model,
+            "prompt": f"<s>[INST] <<SYS>>{self.system_prompt}<</SYS>>{conversation_prompt} [/INST]",
+            "self.stop_sequence": self.stop_sequences,
+            "stream": stream,
+            "top_k": self.top_k,
+            "top_p": self.top_p,
+        }
+
+        def for_stream():
+            response = self.session.post(
+                self.chat_endpoint, json=payload, stream=True, timeout=self.timeout
+            )
+            if (
+                not response.ok
+                or not response.headers.get("Content-Type")
+                == "text/event-stream; charset=utf-8"
+            ):
+                raise Exception(
+                    f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
+                )
+
+            for value in response.iter_lines(
+                decode_unicode=True,
+                delimiter="" if raw else "data:",
+                chunk_size=self.stream_chunk_size,
+            ):
+                try:
+                    resp = json.loads(value)
+                    self.last_response.update(resp)
+                    yield value if raw else resp
+                except json.decoder.JSONDecodeError:
+                    pass
+            self.conversation.update_chat_history(
+                prompt, self.get_message(self.last_response)
+            )
+
+        def for_non_stream():
+            response = self.session.post(
+                self.chat_endpoint, json=payload, stream=False, timeout=self.timeout
+            )
+            if (
+                not response.ok
+                or not response.headers.get("Content-Type", "") == "application/json"
+            ):
+                raise Exception(
+                    f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
+                )
+            resp = response.json()
+            self.last_response.update(resp)
+            self.conversation.update_chat_history(
+                prompt, self.get_message(self.last_response)
+            )
+            return resp
+
+        return for_stream() if stream else for_non_stream()
+
+    def chat(
+        self,
+        prompt: str,
+        stream: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> str:
+        """Generate response `str`
+        Args:
+            prompt (str): Prompt to be send.
+            stream (bool, optional): Flag for streaming response. Defaults to False.
+            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+        Returns:
+            str: Response generated
+        """
+
+        def for_stream():
+            for response in self.ask(
+                prompt, True, optimizer=optimizer, conversationally=conversationally
+            ):
+                yield self.get_message(response)
+
+        def for_non_stream():
+            return self.get_message(
+                self.ask(
+                    prompt,
+                    False,
+                    optimizer=optimizer,
+                    conversationally=conversationally,
+                )
+            )
+
+        return for_stream() if stream else for_non_stream()
+
+    def get_message(self, response: dict) -> str:
+        """Retrieves message only from response
+
+        Args:
+            response (dict): Response generated by `self.ask`
+
+        Returns:
+            str: Message extracted
+        """
+        assert isinstance(response, dict), "Response should be of dict data-type only"
+        return response.get("completion")
+class AsyncLEO(AsyncProvider):
+    def __init__(
+        self,
+        is_conversation: bool = True,
+        max_tokens: int = 600,
+        temperature: float = 0.2,
+        top_k: int = -1,
+        top_p: float = 0.999,
+        model: str = "llama-2-13b-chat",
+        brave_key: str = "qztbjzBqJueQZLFkwTTJrieu8Vw3789u",
+        timeout: int = 30,
+        intro: str = None,
+        filepath: str = None,
+        update_file: bool = True,
+        proxies: dict = {},
+        history_offset: int = 10250,
+        act: str = None,
+    ):
+        """Instantiate TGPT
+
+        Args:
+            is_conversation (str, optional): Flag for chatting conversationally. Defaults to True.
+            brave_key (str, optional): Brave API access key. Defaults to "qztbjzBqJueQZLFkwTTJrieu8Vw3789u".
+            model (str, optional): Text generation model name. Defaults to "llama-2-13b-chat".
+            max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 600.
+            temperature (float, optional): Charge of the generated text's randomness. Defaults to 0.2.
+            top_k (int, optional): Chance of topic being repeated. Defaults to -1.
+            top_p (float, optional): Sampling threshold during inference time. Defaults to 0.999.
+            timeput (int, optional): Http requesting timeout. Defaults to 30
+            intro (str, optional): Conversation introductory prompt. Defaults to `Conversation.intro`.
+            filepath (str, optional): Path to file containing conversation history. Defaults to None.
+            update_file (bool, optional): Add new prompts and responses to the file. Defaults to True.
+            proxies (dict, optional) : Http reqiuest proxies (socks). Defaults to {}.
+            history_offset (int, optional): Limit conversation history to this number of last texts. Defaults to 10250.
+            act (str|int, optional): Awesome prompt key or index. (Used as intro). Defaults to None.
+        """
+        self.is_conversation = is_conversation
+        self.max_tokens_to_sample = max_tokens
+        self.model = model
+        self.stop_sequences = ["</response>", "</s>"]
+        self.temperature = temperature
+        self.top_k = top_k
+        self.top_p = top_p
+        self.chat_endpoint = "https://ai-chat.bsg.brave.com/v1/complete"
+        self.stream_chunk_size = 64
+        self.timeout = timeout
+        self.last_response = {}
+        self.headers = {
+            "Content-Type": "application/json",
+            "accept": "text/event-stream",
+            "x-brave-key": brave_key,
+            "accept-language": "en-US,en;q=0.9",
+            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:99.0) Gecko/20100101 Firefox/110.0",
+        }
+        self.__available_optimizers = (
+            method
+            for method in dir(Optimizers)
+            if callable(getattr(Optimizers, method)) and not method.startswith("__")
+        )
+        Conversation.intro = (
+            AwesomePrompts().get_act(
+                act, raise_not_found=True, default=None, case_insensitive=True
+            )
+            if act
+            else intro or Conversation.intro
+        )
+        self.conversation = Conversation(
+            is_conversation, self.max_tokens_to_sample, filepath, update_file
+        )
+        self.conversation.history_offset = history_offset
+        self.system_prompt = (
+            "\n\nYour name is Leo, a helpful"
+            "respectful and honest AI assistant created by the company Brave. You will be replying to a user of the Brave browser. "
+            "Always respond in a neutral tone. Be polite and courteous. Answer concisely in no more than 50-80 words."
+            "\n\nPlease ensure that your responses are socially unbiased and positive in nature."
+            "If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. "
+            "If you don't know the answer to a question, please don't share false information.\n"
+        )
+        self.session = httpx.AsyncClient(headers=self.headers, proxies=proxies)
+
+    async def ask(
+        self,
+        prompt: str,
+        stream: bool = False,
+        raw: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> dict | AsyncGenerator:
+        """Chat with AI asynchronously.
+
+        Args:
+            prompt (str): Prompt to be send.
+            stream (bool, optional): Flag for streaming response. Defaults to False.
+            raw (bool, optional): Stream back raw response as received. Defaults to False.
+            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+        Returns:
+           dict|AsyncGenerator : ai content
+        ```json
+        {
+            "completion": "\nNext: domestic cat breeds with short hair >>",
+            "stop_reason": null,
+            "truncated": false,
+            "stop": null,
+            "model": "llama-2-13b-chat",
+            "log_id": "cmpl-3kYiYxSNDvgMShSzFooz6t",
+            "exception": null
+        }
+        ```
+        """
+        conversation_prompt = self.conversation.gen_complete_prompt(prompt)
+        if optimizer:
+            if optimizer in self.__available_optimizers:
+                conversation_prompt = getattr(Optimizers, optimizer)(
+                    conversation_prompt if conversationally else prompt
+                )
+            else:
+                raise Exception(
+                    f"Optimizer is not one of {self.__available_optimizers}"
+                )
+
+        payload = {
+            "max_tokens_to_sample": self.max_tokens_to_sample,
+            "model": self.model,
+            "prompt": f"<s>[INST] <<SYS>>{self.system_prompt}<</SYS>>{conversation_prompt} [/INST]",
+            "self.stop_sequence": self.stop_sequences,
+            "stream": stream,
+            "top_k": self.top_k,
+            "top_p": self.top_p,
+        }
+
+        async def for_stream():
+            async with self.session.stream(
+                "POST", self.chat_endpoint, json=payload, timeout=self.timeout
+            ) as response:
+                if (
+                    not response.is_success
+                    or not response.headers.get("Content-Type")
+                    == "text/event-stream; charset=utf-8"
+                ):
+                    raise exceptions.FailedToGenerateResponseError(
+                        f"Failed to generate response - ({response.status_code}, {response.reason_phrase})"
+                    )
+                async for value in response.aiter_lines():
+                    try:
+                        resp = sanitize_stream(value)
+                        self.last_response.update(resp)
+                        yield value if raw else resp
+                    except json.decoder.JSONDecodeError:
+                        pass
+
+            self.conversation.update_chat_history(
+                prompt, await self.get_message(self.last_response)
+            )
+
+        async def for_non_stream():
+            async for _ in for_stream():
+                pass
+            return self.last_response
+
+        return for_stream() if stream else await for_non_stream()
+
+    async def chat(
+        self,
+        prompt: str,
+        stream: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> str | AsyncGenerator:
+        """Generate response `str` asynchronously.
+        Args:
+            prompt (str): Prompt to be send.
+            stream (bool, optional): Flag for streaming response. Defaults to False.
+            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+        Returns:
+            str|AsyncGenerator: Response generated
+        """
+
+        async def for_stream():
+            async_ask = await self.ask(
+                prompt, True, optimizer=optimizer, conversationally=conversationally
+            )
+            async for response in async_ask:
+                yield await self.get_message(response)
+
+        async def for_non_stream():
+            return await self.get_message(
+                await self.ask(
+                    prompt,
+                    False,
+                    optimizer=optimizer,
+                    conversationally=conversationally,
+                )
+            )
+
+        return for_stream() if stream else await for_non_stream()
+
+    async def get_message(self, response: dict) -> str:
+        """Retrieves message only from response
+
+        Args:
+            response (dict): Response generated by `self.ask`
+
+        Returns:
+            str: Message extracted
+        """
+        assert isinstance(response, dict), "Response should be of dict data-type only"
+        return response.get("completion")
+#------------------------------------------------------KOBOLDAI-----------------------------------------------------------
+class KOBOLDAI(Provider):
+    def __init__(
+        self,
+        is_conversation: bool = True,
+        max_tokens: int = 600,
+        temperature: float = 1,
+        top_p: float = 1,
+        timeout: int = 30,
+        intro: str = None,
+        filepath: str = None,
+        update_file: bool = True,
+        proxies: dict = {},
+        history_offset: int = 10250,
+        act: str = None,
+    ):
+        """Instantiate TGPT
+
+        Args:
+            is_conversation (str, optional): Flag for chatting conversationally. Defaults to True.
+            max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 600.
+            temperature (float, optional): Charge of the generated text's randomness. Defaults to 0.2.
+            top_p (float, optional): Sampling threshold during inference time. Defaults to 0.999.
+            timeout (int, optional): Http requesting timeout. Defaults to 30
+            intro (str, optional): Conversation introductory prompt. Defaults to `Conversation.intro`.
+            filepath (str, optional): Path to file containing conversation history. Defaults to None.
+            update_file (bool, optional): Add new prompts and responses to the file. Defaults to True.
+            proxies (dict, optional) : Http reqiuest proxies (socks). Defaults to {}.
+            history_offset (int, optional): Limit conversation history to this number of last texts. Defaults to 10250.
+            act (str|int, optional): Awesome prompt key or index. (Used as intro). Defaults to None.
+        """
+        self.session = requests.Session()
+        self.is_conversation = is_conversation
+        self.max_tokens_to_sample = max_tokens
+        self.temperature = temperature
+        self.top_p = top_p
+        self.chat_endpoint = (
+            "https://koboldai-koboldcpp-tiefighter.hf.space/api/extra/generate/stream"
+        )
+        self.stream_chunk_size = 64
+        self.timeout = timeout
+        self.last_response = {}
+        self.headers = {
+            "Content-Type": "application/json",
+            "Accept": "application/json",
+        }
+
+        self.__available_optimizers = (
+            method
+            for method in dir(Optimizers)
+            if callable(getattr(Optimizers, method)) and not method.startswith("__")
+        )
+        self.session.headers.update(self.headers)
+        Conversation.intro = (
+            AwesomePrompts().get_act(
+                act, raise_not_found=True, default=None, case_insensitive=True
+            )
+            if act
+            else intro or Conversation.intro
+        )
+        self.conversation = Conversation(
+            is_conversation, self.max_tokens_to_sample, filepath, update_file
+        )
+        self.conversation.history_offset = history_offset
+        self.session.proxies = proxies
+
+    def ask(
+        self,
+        prompt: str,
+        stream: bool = False,
+        raw: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> dict:
+        """Chat with AI
+
+        Args:
+            prompt (str): Prompt to be send.
+            stream (bool, optional): Flag for streaming response. Defaults to False.
+            raw (bool, optional): Stream back raw response as received. Defaults to False.
+            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+        Returns:
+           dict : {}
+        ```json
+        {
+           "token" : "How may I assist you today?"
+        }
+        ```
+        """
+        conversation_prompt = self.conversation.gen_complete_prompt(prompt)
+        if optimizer:
+            if optimizer in self.__available_optimizers:
+                conversation_prompt = getattr(Optimizers, optimizer)(
+                    conversation_prompt if conversationally else prompt
+                )
+            else:
+                raise Exception(
+                    f"Optimizer is not one of {self.__available_optimizers}"
+                )
+
+        self.session.headers.update(self.headers)
+        payload = {
+            "prompt": conversation_prompt,
+            "temperature": self.temperature,
+            "top_p": self.top_p,
+        }
+
+        def for_stream():
+            response = self.session.post(
+                self.chat_endpoint, json=payload, stream=True, timeout=self.timeout
+            )
+            if not response.ok:
+                raise Exception(
+                    f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
+                )
+
+            message_load = ""
+            for value in response.iter_lines(
+                decode_unicode=True,
+                delimiter="" if raw else "event: message\ndata:",
+                chunk_size=self.stream_chunk_size,
+            ):
+                try:
+                    resp = json.loads(value)
+                    message_load += self.get_message(resp)
+                    resp["token"] = message_load
+                    self.last_response.update(resp)
+                    yield value if raw else resp
+                except json.decoder.JSONDecodeError:
+                    pass
+            self.conversation.update_chat_history(
+                prompt, self.get_message(self.last_response)
+            )
+
+        def for_non_stream():
+            # let's make use of stream
+            for _ in for_stream():
+                pass
+            return self.last_response
+
+        return for_stream() if stream else for_non_stream()
+
+    def chat(
+        self,
+        prompt: str,
+        stream: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> str:
+        """Generate response `str`
+        Args:
+            prompt (str): Prompt to be send.
+            stream (bool, optional): Flag for streaming response. Defaults to False.
+            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+        Returns:
+            str: Response generated
+        """
+
+        def for_stream():
+            for response in self.ask(
+                prompt, True, optimizer=optimizer, conversationally=conversationally
+            ):
+                yield self.get_message(response)
+
+        def for_non_stream():
+            return self.get_message(
+                self.ask(
+                    prompt,
+                    False,
+                    optimizer=optimizer,
+                    conversationally=conversationally,
+                )
+            )
+
+        return for_stream() if stream else for_non_stream()
+
+    def get_message(self, response: dict) -> str:
+        """Retrieves message only from response
+
+        Args:
+            response (dict): Response generated by `self.ask`
+
+        Returns:
+            str: Message extracted
+        """
+        assert isinstance(response, dict), "Response should be of dict data-type only"
+        return response.get("token")
+class AsyncKOBOLDAI(AsyncProvider):
+    def __init__(
+        self,
+        is_conversation: bool = True,
+        max_tokens: int = 600,
+        temperature: float = 1,
+        top_p: float = 1,
+        timeout: int = 30,
+        intro: str = None,
+        filepath: str = None,
+        update_file: bool = True,
+        proxies: dict = {},
+        history_offset: int = 10250,
+        act: str = None,
+    ):
+        """Instantiate TGPT
+
+        Args:
+            is_conversation (str, optional): Flag for chatting conversationally. Defaults to True.
+            max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 600.
+            temperature (float, optional): Charge of the generated text's randomness. Defaults to 0.2.
+            top_p (float, optional): Sampling threshold during inference time. Defaults to 0.999.
+            timeout (int, optional): Http requesting timeout. Defaults to 30
+            intro (str, optional): Conversation introductory prompt. Defaults to `Conversation.intro`.
+            filepath (str, optional): Path to file containing conversation history. Defaults to None.
+            update_file (bool, optional): Add new prompts and responses to the file. Defaults to True.
+            proxies (dict, optional) : Http reqiuest proxies (socks). Defaults to {}.
+            history_offset (int, optional): Limit conversation history to this number of last texts. Defaults to 10250.
+            act (str|int, optional): Awesome prompt key or index. (Used as intro). Defaults to None.
+        """
+        self.is_conversation = is_conversation
+        self.max_tokens_to_sample = max_tokens
+        self.temperature = temperature
+        self.top_p = top_p
+        self.chat_endpoint = (
+            "https://koboldai-koboldcpp-tiefighter.hf.space/api/extra/generate/stream"
+        )
+        self.stream_chunk_size = 64
+        self.timeout = timeout
+        self.last_response = {}
+        self.headers = {
+            "Content-Type": "application/json",
+            "Accept": "application/json",
+        }
+
+        self.__available_optimizers = (
+            method
+            for method in dir(Optimizers)
+            if callable(getattr(Optimizers, method)) and not method.startswith("__")
+        )
+        Conversation.intro = (
+            AwesomePrompts().get_act(
+                act, raise_not_found=True, default=None, case_insensitive=True
+            )
+            if act
+            else intro or Conversation.intro
+        )
+        self.conversation = Conversation(
+            is_conversation, self.max_tokens_to_sample, filepath, update_file
+        )
+        self.conversation.history_offset = history_offset
+        self.session = httpx.AsyncClient(headers=self.headers, proxies=proxies)
+
+    async def ask(
+        self,
+        prompt: str,
+        stream: bool = False,
+        raw: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> dict | AsyncGenerator:
+        """Chat with AI asynchronously.
+
+        Args:
+            prompt (str): Prompt to be send.
+            stream (bool, optional): Flag for streaming response. Defaults to False.
+            raw (bool, optional): Stream back raw response as received. Defaults to False.
+            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+        Returns:
+           dict|AsyncGenerator : ai content
+        ```json
+        {
+           "token" : "How may I assist you today?"
+        }
+        ```
+        """
+        conversation_prompt = self.conversation.gen_complete_prompt(prompt)
+        if optimizer:
+            if optimizer in self.__available_optimizers:
+                conversation_prompt = getattr(Optimizers, optimizer)(
+                    conversation_prompt if conversationally else prompt
+                )
+            else:
+                raise Exception(
+                    f"Optimizer is not one of {self.__available_optimizers}"
+                )
+
+        payload = {
+            "prompt": conversation_prompt,
+            "temperature": self.temperature,
+            "top_p": self.top_p,
+        }
+
+        async def for_stream():
+            async with self.session.stream(
+                "POST", self.chat_endpoint, json=payload, timeout=self.timeout
+            ) as response:
+                if not response.is_success:
+                    raise exceptions.FailedToGenerateResponseError(
+                        f"Failed to generate response - ({response.status_code}, {response.reason_phrase})"
+                    )
+
+                message_load = ""
+                async for value in response.aiter_lines():
+                    try:
+                        resp = sanitize_stream(value)
+                        message_load += await self.get_message(resp)
+                        resp["token"] = message_load
+                        self.last_response.update(resp)
+                        yield value if raw else resp
+                    except json.decoder.JSONDecodeError:
+                        pass
+
+            self.conversation.update_chat_history(
+                prompt, await self.get_message(self.last_response)
+            )
+
+        async def for_non_stream():
+            # let's make use of stream
+            async for _ in for_stream():
+                pass
+            return self.last_response
+
+        return for_stream() if stream else await for_non_stream()
+
+    async def chat(
+        self,
+        prompt: str,
+        stream: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> str | AsyncGenerator:
+        """Generate response `str` asynchronously.
+        Args:
+            prompt (str): Prompt to be send.
+            stream (bool, optional): Flag for streaming response. Defaults to False.
+            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+        Returns:
+            str: Response generated
+        """
+
+        async def for_stream():
+            async_ask = await self.ask(
+                prompt, True, optimizer=optimizer, conversationally=conversationally
+            )
+            async for response in async_ask:
+                yield await self.get_message(response)
+
+        async def for_non_stream():
+            return await self.get_message(
+                await self.ask(
+                    prompt,
+                    False,
+                    optimizer=optimizer,
+                    conversationally=conversationally,
+                )
+            )
+
+        return for_stream() if stream else await for_non_stream()
+
+    async def get_message(self, response: dict) -> str:
+        """Retrieves message only from response
+
+        Args:
+            response (dict): Response generated by `self.ask`
+
+        Returns:
+            str: Message extracted
+        """
+        assert isinstance(response, dict), "Response should be of dict data-type only"
+        return response.get("token")
+#------------------------------------------------------OpenGPT-----------------------------------------------------------
+class OPENGPT:
+    def __init__(
+        self,
+        is_conversation: bool = True,
+        max_tokens: int = 600,
+        timeout: int = 30,
+        intro: str = None,
+        filepath: str = None,
+        update_file: bool = True,
+        proxies: dict = {},
+        history_offset: int = 10250,
+        act: str = None,
+    ):
+        """Instantiates OPENGPT
+
+        Args:
+            is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True
+            max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 600.
+            timeout (int, optional): Http request timeout. Defaults to 30.
+            intro (str, optional): Conversation introductory prompt. Defaults to None.
+            filepath (str, optional): Path to file containing conversation history. Defaults to None.
+            update_file (bool, optional): Add new prompts and responses to the file. Defaults to True.
+            proxies (dict, optional): Http request proxies. Defaults to {}.
+            history_offset (int, optional): Limit conversation history to this number of last texts. Defaults to 10250.
+            act (str|int, optional): Awesome prompt key or index. (Used as intro). Defaults to None.
+        """
+        self.session = requests.Session()
+        self.max_tokens_to_sample = max_tokens
+        self.is_conversation = is_conversation
+        self.chat_endpoint = (
+            "https://opengpts-example-vz4y4ooboq-uc.a.run.app/runs/stream"
+        )
+        self.stream_chunk_size = 64
+        self.timeout = timeout
+        self.last_response = {}
+        self.assistant_id = "bca37014-6f97-4f2b-8928-81ea8d478d88"
+        self.authority = "opengpts-example-vz4y4ooboq-uc.a.run.app"
+
+        self.headers = {
+            "authority": self.authority,
+            "accept": "text/event-stream",
+            "accept-language": "en-US,en;q=0.7",
+            "cache-control": "no-cache",
+            "content-type": "application/json",
+            "origin": "https://opengpts-example-vz4y4ooboq-uc.a.run.app",
+            "pragma": "no-cache",
+            "referer": "https://opengpts-example-vz4y4ooboq-uc.a.run.app/",
+            "sec-fetch-site": "same-origin",
+            "sec-gpc": "1",
+            "user-agent": "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36",
+        }
+
+        self.__available_optimizers = (
+            method
+            for method in dir(Optimizers)
+            if callable(getattr(Optimizers, method)) and not method.startswith("__")
+        )
+        self.session.headers.update(self.headers)
+        Conversation.intro = (
+            AwesomePrompts().get_act(
+                act, raise_not_found=True, default=None, case_insensitive=True
+            )
+            if act
+            else intro or Conversation.intro
+        )
+        self.conversation = Conversation(
+            is_conversation, self.max_tokens_to_sample, filepath, update_file
+        )
+        self.conversation.history_offset = history_offset
+        self.session.proxies = proxies
+
+    def ask(
+        self,
+        prompt: str,
+        stream: bool = False,
+        raw: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> dict:
+        """Chat with AI
+
+        Args:
+            prompt (str): Prompt to be send.
+            stream (bool, optional): Flag for streaming response. Defaults to False.
+            raw (bool, optional): Stream back raw response as received. Defaults to False.
+            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+        Returns:
+           dict : {}
+        ```json
+        {
+            "messages": [
+                {
+                    "content": "Hello there",
+                    "additional_kwargs": {},
+                    "type": "human",
+                    "example": false
+                },
+                {
+                    "content": "Hello! How can I assist you today?",
+                    "additional_kwargs": {
+                    "agent": {
+                        "return_values": {
+                            "output": "Hello! How can I assist you today?"
+                            },
+                        "log": "Hello! How can I assist you today?",
+                        "type": "AgentFinish"
+                    }
+                },
+                "type": "ai",
+                "example": false
+                }]
+        }
+        ```
+        """
+        conversation_prompt = self.conversation.gen_complete_prompt(prompt)
+        if optimizer:
+            if optimizer in self.__available_optimizers:
+                conversation_prompt = getattr(Optimizers, optimizer)(
+                    conversation_prompt if conversationally else prompt
+                )
+            else:
+                raise Exception(
+                    f"Optimizer is not one of {self.__available_optimizers}"
+                )
+
+        self.session.headers.update(self.headers)
+        self.session.headers.update(
+            dict(
+                cookie=f"opengpts_user_id={uuid4().__str__()}",
+            )
+        )
+        payload = {
+            "input": [
+                {
+                    "content": conversation_prompt,
+                    "additional_kwargs": {},
+                    "type": "human",
+                    "example": False,
+                },
+            ],
+            "assistant_id": self.assistant_id,
+            "thread_id": "",
+        }
+
+        def for_stream():
+            response = self.session.post(
+                self.chat_endpoint, json=payload, stream=True, timeout=self.timeout
+            )
+            if (
+                not response.ok
+                or not response.headers.get("Content-Type")
+                == "text/event-stream; charset=utf-8"
+            ):
+                raise Exception(
+                    f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
+                )
+
+            for value in response.iter_lines(
+                decode_unicode=True,
+                chunk_size=self.stream_chunk_size,
+            ):
+                try:
+                    modified_value = re.sub("data:", "", value)
+                    resp = json.loads(modified_value)
+                    if len(resp) == 1:
+                        continue
+                    self.last_response.update(resp[1])
+                    yield value if raw else resp[1]
+                except json.decoder.JSONDecodeError:
+                    pass
+            self.conversation.update_chat_history(
+                prompt, self.get_message(self.last_response)
+            )
+
+        def for_non_stream():
+            for _ in for_stream():
+                pass
+            return self.last_response
+
+        return for_stream() if stream else for_non_stream()
+
+    def chat(
+        self,
+        prompt: str,
+        stream: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> str:
+        """Generate response `str`
+        Args:
+            prompt (str): Prompt to be send.
+            stream (bool, optional): Flag for streaming response. Defaults to False.
+            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+        Returns:
+            str: Response generated
+        """
+
+        def for_stream():
+            for response in self.ask(
+                prompt, True, optimizer=optimizer, conversationally=conversationally
+            ):
+                yield self.get_message(response)
+
+        def for_non_stream():
+            return self.get_message(
+                self.ask(
+                    prompt,
+                    False,
+                    optimizer=optimizer,
+                    conversationally=conversationally,
+                )
+            )
+
+        return for_stream() if stream else for_non_stream()
+
+    def get_message(self, response: dict) -> str:
+        """Retrieves message only from response
+
+        Args:
+            response (dict): Response generated by `self.ask`
+
+        Returns:
+            str: Message extracted
+        """
+        assert isinstance(response, dict), "Response should be of dict data-type only"
+        return response["content"]
+class AsyncOPENGPT(AsyncProvider):
+    def __init__(
+        self,
+        is_conversation: bool = True,
+        max_tokens: int = 600,
+        timeout: int = 30,
+        intro: str = None,
+        filepath: str = None,
+        update_file: bool = True,
+        proxies: dict = {},
+        history_offset: int = 10250,
+        act: str = None,
+    ):
+        """Instantiates OPENGPT
+
+        Args:
+            is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True
+            max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 600.
+            timeout (int, optional): Http request timeout. Defaults to 30.
+            intro (str, optional): Conversation introductory prompt. Defaults to None.
+            filepath (str, optional): Path to file containing conversation history. Defaults to None.
+            update_file (bool, optional): Add new prompts and responses to the file. Defaults to True.
+            proxies (dict, optional): Http request proxies. Defaults to {}.
+            history_offset (int, optional): Limit conversation history to this number of last texts. Defaults to 10250.
+            act (str|int, optional): Awesome prompt key or index. (Used as intro). Defaults to None.
+        """
+        self.max_tokens_to_sample = max_tokens
+        self.is_conversation = is_conversation
+        self.chat_endpoint = (
+            "https://opengpts-example-vz4y4ooboq-uc.a.run.app/runs/stream"
+        )
+        self.stream_chunk_size = 64
+        self.timeout = timeout
+        self.last_response = {}
+        self.assistant_id = "bca37014-6f97-4f2b-8928-81ea8d478d88"
+        self.authority = "opengpts-example-vz4y4ooboq-uc.a.run.app"
+
+        self.headers = {
+            "authority": self.authority,
+            "accept": "text/event-stream",
+            "accept-language": "en-US,en;q=0.7",
+            "cache-control": "no-cache",
+            "content-type": "application/json",
+            "origin": "https://opengpts-example-vz4y4ooboq-uc.a.run.app",
+            "pragma": "no-cache",
+            "referer": "https://opengpts-example-vz4y4ooboq-uc.a.run.app/",
+            "sec-fetch-site": "same-origin",
+            "sec-gpc": "1",
+            "user-agent": "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36",
+        }
+
+        self.__available_optimizers = (
+            method
+            for method in dir(Optimizers)
+            if callable(getattr(Optimizers, method)) and not method.startswith("__")
+        )
+        Conversation.intro = (
+            AwesomePrompts().get_act(
+                act, raise_not_found=True, default=None, case_insensitive=True
+            )
+            if act
+            else intro or Conversation.intro
+        )
+        self.conversation = Conversation(
+            is_conversation, self.max_tokens_to_sample, filepath, update_file
+        )
+        self.conversation.history_offset = history_offset
+        self.session = httpx.AsyncClient(headers=self.headers, proxies=proxies)
+
+    async def ask(
+        self,
+        prompt: str,
+        stream: bool = False,
+        raw: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> dict | AsyncGenerator:
+        """Chat with AI asynchronously
+
+        Args:
+            prompt (str): Prompt to be send.
+            stream (bool, optional): Flag for streaming response. Defaults to False.
+            raw (bool, optional): Stream back raw response as received. Defaults to False.
+            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+        Returns:
+           dict|AsyncGenerator : ai content.
+        ```json
+        {
+            "messages": [
+                {
+                    "content": "Hello there",
+                    "additional_kwargs": {},
+                    "type": "human",
+                    "example": false
+                },
+                {
+                    "content": "Hello! How can I assist you today?",
+                    "additional_kwargs": {
+                    "agent": {
+                        "return_values": {
+                            "output": "Hello! How can I assist you today?"
+                            },
+                        "log": "Hello! How can I assist you today?",
+                        "type": "AgentFinish"
+                    }
+                },
+                "type": "ai",
+                "example": false
+                }]
+        }
+        ```
+        """
+        conversation_prompt = self.conversation.gen_complete_prompt(prompt)
+        if optimizer:
+            if optimizer in self.__available_optimizers:
+                conversation_prompt = getattr(Optimizers, optimizer)(
+                    conversation_prompt if conversationally else prompt
+                )
+            else:
+                raise Exception(
+                    f"Optimizer is not one of {self.__available_optimizers}"
+                )
+        self.headers.update(
+            dict(
+                cookie=f"opengpts_user_id={uuid4().__str__()}",
+            )
+        )
+        payload = {
+            "input": [
+                {
+                    "content": conversation_prompt,
+                    "additional_kwargs": {},
+                    "type": "human",
+                    "example": False,
+                },
+            ],
+            "assistant_id": self.assistant_id,
+            "thread_id": "",
+        }
+
+        async def for_stream():
+            async with self.session.stream(
+                "POST",
+                self.chat_endpoint,
+                json=payload,
+                timeout=self.timeout,
+                headers=self.headers,
+            ) as response:
+                if (
+                    not response.is_success
+                    or not response.headers.get("Content-Type")
+                    == "text/event-stream; charset=utf-8"
+                ):
+                    raise exceptions.FailedToGenerateResponseError(
+                        f"Failed to generate response - ({response.status_code}, {response.reason_phrase}) - {response.text}"
+                    )
+
+                async for value in response.aiter_lines():
+                    try:
+                        modified_value = re.sub("data:", "", value)
+                        resp = json.loads(modified_value)
+                        if len(resp) == 1:
+                            continue
+                        self.last_response.update(resp[1])
+                        yield value if raw else resp[1]
+                    except json.decoder.JSONDecodeError:
+                        pass
+
+            self.conversation.update_chat_history(
+                prompt, await self.get_message(self.last_response)
+            )
+
+        async def for_non_stream():
+            async for _ in for_stream():
+                pass
+            return self.last_response
+
+        return for_stream() if stream else await for_non_stream()
+
+    async def chat(
+        self,
+        prompt: str,
+        stream: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> str | AsyncGenerator:
+        """Generate response `str` asynchronously.
+        Args:
+            prompt (str): Prompt to be send.
+            stream (bool, optional): Flag for streaming response. Defaults to False.
+            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+        Returns:
+            str|AsyncGenerator: Response generated
+        """
+
+        async def for_stream():
+            async_ask = await self.ask(
+                prompt, True, optimizer=optimizer, conversationally=conversationally
+            )
+            async for response in async_ask:
+                yield await self.get_message(response)
+
+        async def for_non_stream():
+            return await self.get_message(
+                await self.ask(
+                    prompt,
+                    False,
+                    optimizer=optimizer,
+                    conversationally=conversationally,
+                )
+            )
+
+        return for_stream() if stream else await for_non_stream()
+
+    async def get_message(self, response: dict) -> str:
+        """Retrieves message only from response
+
+        Args:
+            response (dict): Response generated by `self.ask`
+
+        Returns:
+            str: Message extracted
+        """
+        assert isinstance(response, dict), "Response should be of dict data-type only"
+        return response["content"]
+#------------------------------------------------------PERPLEXITY--------------------------------------------------------  
+class PERPLEXITY(Provider):
+    def __init__(
+        self,
+        is_conversation: bool = True,
+        max_tokens: int = 600,
+        timeout: int = 30,
+        intro: str = None,
+        filepath: str = None,
+        update_file: bool = True,
+        proxies: dict = {},
+        history_offset: int = 10250,
+        act: str = None,
+        quiet: bool = False,
+    ):
+        """Instantiates PERPLEXITY
+
+        Args:
+            is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True
+            max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 600.
+            timeout (int, optional): Http request timeout. Defaults to 30.
+            intro (str, optional): Conversation introductory prompt. Defaults to None.
+            filepath (str, optional): Path to file containing conversation history. Defaults to None.
+            update_file (bool, optional): Add new prompts and responses to the file. Defaults to True.
+            proxies (dict, optional): Http request proxies. Defaults to {}.
+            history_offset (int, optional): Limit conversation history to this number of last texts. Defaults to 10250.
+            act (str|int, optional): Awesome prompt key or index. (Used as intro). Defaults to None.
+            quiet (bool, optional): Ignore web search-results and yield final response only. Defaults to False.
+        """
+        self.max_tokens_to_sample = max_tokens
+        self.is_conversation = is_conversation
+        self.last_response = {}
+        self.web_results: dict = {}
+        self.quiet = quiet
+
+        self.__available_optimizers = (
+            method
+            for method in dir(Optimizers)
+            if callable(getattr(Optimizers, method)) and not method.startswith("__")
+        )
+        Conversation.intro = (
+            AwesomePrompts().get_act(
+                act, raise_not_found=True, default=None, case_insensitive=True
+            )
+            if act
+            else intro or Conversation.intro
+        )
+        self.conversation = Conversation(
+            is_conversation, self.max_tokens_to_sample, filepath, update_file
+        )
+        self.conversation.history_offset = history_offset
+
+    def ask(
+        self,
+        prompt: str,
+        stream: bool = False,
+        raw: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> dict:
+        """Chat with AI
+
+                Args:
+                    prompt (str): Prompt to be send.
+                    stream (bool, optional): Flag for streaming response. Defaults to False.
+                    raw (bool, optional): Stream back raw response as received. Defaults to False.
+                    optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+                    conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+                Returns:
+                   dict : {}
+                ```json
+        {
+            "status": "pending",
+            "uuid": "3604dfcc-611f-4b7d-989d-edca2a7233c7",
+            "read_write_token": null,
+            "frontend_context_uuid": "f6d43119-5231-481d-b692-f52e1f52d2c6",
+            "final": false,
+            "backend_uuid": "a6d6ec9e-da69-4841-af74-0de0409267a8",
+            "media_items": [],
+            "widget_data": [],
+            "knowledge_cards": [],
+            "expect_search_results": "false",
+            "mode": "concise",
+            "search_focus": "internet",
+            "gpt4": false,
+            "display_model": "turbo",
+            "attachments": null,
+            "answer": "",
+            "web_results": [],
+            "chunks": [],
+            "extra_web_results": []
+        }
+                ```
+        """
+        conversation_prompt = self.conversation.gen_complete_prompt(prompt)
+        if optimizer:
+            if optimizer in self.__available_optimizers:
+                conversation_prompt = getattr(Optimizers, optimizer)(
+                    conversation_prompt if conversationally else prompt
+                )
+            else:
+                raise Exception(
+                    f"Optimizer is not one of {self.__available_optimizers}"
+                )
+
+        def for_stream():
+            for response in Perplexity().generate_answer(conversation_prompt):
+                yield json.dumps(response) if raw else response
+                self.last_response.update(response)
+
+            self.conversation.update_chat_history(
+                prompt,
+                self.get_message(self.last_response),
+            )
+
+        def for_non_stream():
+            for _ in for_stream():
+                pass
+            return self.last_response
+
+        return for_stream() if stream else for_non_stream()
+
+    def chat(
+        self,
+        prompt: str,
+        stream: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> str:
+        """Generate response `str`
+        Args:
+            prompt (str): Prompt to be send.
+            stream (bool, optional): Flag for streaming response. Defaults to False.
+            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+        Returns:
+            str: Response generated
+        """
+
+        def for_stream():
+            for response in self.ask(
+                prompt, True, optimizer=optimizer, conversationally=conversationally
+            ):
+                yield self.get_message(response)
+
+        def for_non_stream():
+            return self.get_message(
+                self.ask(
+                    prompt,
+                    False,
+                    optimizer=optimizer,
+                    conversationally=conversationally,
+                )
+            )
+
+        return for_stream() if stream else for_non_stream()
+
+    def get_message(self, response: dict) -> str:
+        """Retrieves message only from response
+
+        Args:
+            response (dict): Response generated by `self.ask`
+
+        Returns:
+            str: Message extracted
+        """
+        assert isinstance(response, dict), "Response should be of dict data-type only"
+        text_str: str = response.get("answer", "")
+
+        def update_web_results(web_results: list) -> None:
+            for index, results in enumerate(web_results, start=1):
+                self.web_results[str(index) + ". " + results["name"]] = dict(
+                    url=results.get("url"), snippet=results.get("snippet")
+                )
+
+        if response.get("text"):
+            # last chunk
+            target: dict[str, Any] = json.loads(response.get("text"))
+            text_str = target.get("answer")
+            web_results: list[dict] = target.get("web_results")
+            self.web_results.clear()
+            update_web_results(web_results)
+
+            return (
+                text_str
+                if self.quiet or not self.web_results
+                else text_str + "\n\n# WEB-RESULTS\n\n" + yaml.dump(self.web_results)
+            )
+
+        else:
+            if str(response.get("expect_search_results")).lower() == "true":
+                return (
+                    text_str
+                    if self.quiet
+                    else text_str
+                    + "\n\n# WEB-RESULTS\n\n"
+                    + yaml.dump(response.get("web_results"))
+                )
+            else:
+                return text_str
+#------------------------------------------------------BLACKBOXAI--------------------------------------------------------  
+class BLACKBOXAI:
+    def __init__(
+        self,
+        is_conversation: bool = True,
+        max_tokens: int = 8000,
+        timeout: int = 30,
+        intro: str = None,
+        filepath: str = None,
+        update_file: bool = True,
+        proxies: dict = {},
+        history_offset: int = 10250,
+        act: str = None,
+        model: str = None,
+    ):
+        """Instantiates BLACKBOXAI
+
+        Args:
+            is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True
+            max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 600.
+            timeout (int, optional): Http request timeout. Defaults to 30.
+            intro (str, optional): Conversation introductory prompt. Defaults to None.
+            filepath (str, optional): Path to file containing conversation history. Defaults to None.
+            update_file (bool, optional): Add new prompts and responses to the file. Defaults to True.
+            proxies (dict, optional): Http request proxies. Defaults to {}.
+            history_offset (int, optional): Limit conversation history to this number of last texts. Defaults to 10250.
+            act (str|int, optional): Awesome prompt key or index. (Used as intro). Defaults to None.
+            model (str, optional): Model name. Defaults to "Phind Model".
+        """
+        self.session = requests.Session()
+        self.max_tokens_to_sample = max_tokens
+        self.is_conversation = is_conversation
+        self.chat_endpoint = "https://www.blackbox.ai/api/chat"
+        self.stream_chunk_size = 64
+        self.timeout = timeout
+        self.last_response = {}
+        self.model = model
+        self.previewToken: str = None
+        self.userId: str = ""
+        self.codeModelMode: bool = True
+        self.id: str = ""
+        self.agentMode: dict = {}
+        self.trendingAgentMode: dict = {}
+        self.isMicMode: bool = False
+
+        self.headers = {
+            "Content-Type": "application/json",
+            "User-Agent": "",
+            "Accept": "*/*",
+            "Accept-Encoding": "Identity",
+        }
+
+        self.__available_optimizers = (
+            method
+            for method in dir(Optimizers)
+            if callable(getattr(Optimizers, method)) and not method.startswith("__")
+        )
+        self.session.headers.update(self.headers)
+        Conversation.intro = (
+            AwesomePrompts().get_act(
+                act, raise_not_found=True, default=None, case_insensitive=True
+            )
+            if act
+            else intro or Conversation.intro
+        )
+        self.conversation = Conversation(
+            is_conversation, self.max_tokens_to_sample, filepath, update_file
+        )
+        self.conversation.history_offset = history_offset
+        self.session.proxies = proxies
+
+    def ask(
+        self,
+        prompt: str,
+        stream: bool = False,
+        raw: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> dict:
+        """Chat with AI
+
+        Args:
+            prompt (str): Prompt to be send.
+            stream (bool, optional): Flag for streaming response. Defaults to False.
+            raw (bool, optional): Stream back raw response as received. Defaults to False.
+            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+        Returns:
+           dict : {}
+        ```json
+        {
+           "text" : "print('How may I help you today?')"
+        }
+        ```
+        """
+        conversation_prompt = self.conversation.gen_complete_prompt(prompt)
+        if optimizer:
+            if optimizer in self.__available_optimizers:
+                conversation_prompt = getattr(Optimizers, optimizer)(
+                    conversation_prompt if conversationally else prompt
+                )
+            else:
+                raise Exception(
+                    f"Optimizer is not one of {self.__available_optimizers}"
+                )
+
+        self.session.headers.update(self.headers)
+        payload = {
+            "messages": [
+                # json.loads(prev_messages),
+                {"content": conversation_prompt, "role": "user"}
+            ],
+            "id": self.id,
+            "previewToken": self.previewToken,
+            "userId": self.userId,
+            "codeModelMode": self.codeModelMode,
+            "agentMode": self.agentMode,
+            "trendingAgentMode": self.trendingAgentMode,
+            "isMicMode": self.isMicMode,
+        }
+
+        def for_stream():
+            response = self.session.post(
+                self.chat_endpoint, json=payload, stream=True, timeout=self.timeout
+            )
+            if (
+                not response.ok
+                or not response.headers.get("Content-Type")
+                == "text/plain; charset=utf-8"
+            ):
+                raise Exception(
+                    f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
+                )
+            streaming_text = ""
+            for value in response.iter_lines(
+                decode_unicode=True,
+                chunk_size=self.stream_chunk_size,
+                delimiter="\n",
+            ):
+                try:
+                    if bool(value):
+                        streaming_text += value + ("\n" if stream else "")
+
+                        resp = dict(text=streaming_text)
+                        self.last_response.update(resp)
+                        yield value if raw else resp
+                except json.decoder.JSONDecodeError:
+                    pass
+            self.conversation.update_chat_history(
+                prompt, self.get_message(self.last_response)
+            )
+
+        def for_non_stream():
+            for _ in for_stream():
+                pass
+            return self.last_response
+
+        return for_stream() if stream else for_non_stream()
+
+    def chat(
+        self,
+        prompt: str,
+        stream: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> str:
+        """Generate response `str`
+        Args:
+            prompt (str): Prompt to be send.
+            stream (bool, optional): Flag for streaming response. Defaults to False.
+            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+        Returns:
+            str: Response generated
+        """
+
+        def for_stream():
+            for response in self.ask(
+                prompt, True, optimizer=optimizer, conversationally=conversationally
+            ):
+                yield self.get_message(response)
+
+        def for_non_stream():
+            return self.get_message(
+                self.ask(
+                    prompt,
+                    False,
+                    optimizer=optimizer,
+                    conversationally=conversationally,
+                )
+            )
+
+        return for_stream() if stream else for_non_stream()
+
+    def get_message(self, response: dict) -> str:
+        """Retrieves message only from response
+
+        Args:
+            response (dict): Response generated by `self.ask`
+
+        Returns:
+            str: Message extracted
+        """
+        assert isinstance(response, dict), "Response should be of dict data-type only"
+        return response["text"]
+    @staticmethod
+    def chat_cli(prompt):
+        """Sends a request to the BLACKBOXAI API and processes the response."""
+        blackbox_ai = BLACKBOXAI()  # Initialize a BLACKBOXAI instance
+        response = blackbox_ai.ask(prompt)  # Perform a chat with the given prompt
+        processed_response = blackbox_ai.get_message(response)  # Process the response
+        print(processed_response)
+class AsyncBLACKBOXAI(AsyncProvider):
+    def __init__(
+        self,
+        is_conversation: bool = True,
+        max_tokens: int = 600,
+        timeout: int = 30,
+        intro: str = None,
+        filepath: str = None,
+        update_file: bool = True,
+        proxies: dict = {},
+        history_offset: int = 10250,
+        act: str = None,
+        model: str = None,
+    ):
+        """Instantiates BLACKBOXAI
+
+        Args:
+            is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True
+            max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 600.
+            timeout (int, optional): Http request timeout. Defaults to 30.
+            intro (str, optional): Conversation introductory prompt. Defaults to None.
+            filepath (str, optional): Path to file containing conversation history. Defaults to None.
+            update_file (bool, optional): Add new prompts and responses to the file. Defaults to True.
+            proxies (dict, optional): Http request proxies. Defaults to {}.
+            history_offset (int, optional): Limit conversation history to this number of last texts. Defaults to 10250.
+            act (str|int, optional): Awesome prompt key or index. (Used as intro). Defaults to None.
+            model (str, optional): Model name. Defaults to "Phind Model".
+        """
+        self.max_tokens_to_sample = max_tokens
+        self.is_conversation = is_conversation
+        self.chat_endpoint = "https://www.blackbox.ai/api/chat"
+        self.stream_chunk_size = 64
+        self.timeout = timeout
+        self.last_response = {}
+        self.model = model
+        self.previewToken: str = None
+        self.userId: str = ""
+        self.codeModelMode: bool = True
+        self.id: str = ""
+        self.agentMode: dict = {}
+        self.trendingAgentMode: dict = {}
+        self.isMicMode: bool = False
+
+        self.headers = {
+            "Content-Type": "application/json",
+            "User-Agent": "",
+            "Accept": "*/*",
+            "Accept-Encoding": "Identity",
+        }
+
+        self.__available_optimizers = (
+            method
+            for method in dir(Optimizers)
+            if callable(getattr(Optimizers, method)) and not method.startswith("__")
+        )
+        Conversation.intro = (
+            AwesomePrompts().get_act(
+                act, raise_not_found=True, default=None, case_insensitive=True
+            )
+            if act
+            else intro or Conversation.intro
+        )
+        self.conversation = Conversation(
+            is_conversation, self.max_tokens_to_sample, filepath, update_file
+        )
+        self.conversation.history_offset = history_offset
+        self.session = httpx.AsyncClient(headers=self.headers, proxies=proxies)
+
+    async def ask(
+        self,
+        prompt: str,
+        stream: bool = False,
+        raw: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> dict | AsyncGenerator:
+        """Chat with AI asynchronously.
+
+        Args:
+            prompt (str): Prompt to be send.
+            stream (bool, optional): Flag for streaming response. Defaults to False.
+            raw (bool, optional): Stream back raw response as received. Defaults to False.
+            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+        Returns:
+           dict|AsyncGenerator : ai content
+        ```json
+        {
+           "text" : "print('How may I help you today?')"
+        }
+        ```
+        """
+        conversation_prompt = self.conversation.gen_complete_prompt(prompt)
+        if optimizer:
+            if optimizer in self.__available_optimizers:
+                conversation_prompt = getattr(Optimizers, optimizer)(
+                    conversation_prompt if conversationally else prompt
+                )
+            else:
+                raise Exception(
+                    f"Optimizer is not one of {self.__available_optimizers}"
+                )
+
+        payload = {
+            "messages": [
+                # json.loads(prev_messages),
+                {"content": conversation_prompt, "role": "user"}
+            ],
+            "id": self.id,
+            "previewToken": self.previewToken,
+            "userId": self.userId,
+            "codeModelMode": self.codeModelMode,
+            "agentMode": self.agentMode,
+            "trendingAgentMode": self.trendingAgentMode,
+            "isMicMode": self.isMicMode,
+        }
+
+        async def for_stream():
+            async with self.session.stream(
+                "POST", self.chat_endpoint, json=payload, timeout=self.timeout
+            ) as response:
+                if (
+                    not response.is_success
+                    or not response.headers.get("Content-Type")
+                    == "text/plain; charset=utf-8"
+                ):
+                    raise exceptions.FailedToGenerateResponseError(
+                        f"Failed to generate response - ({response.status_code}, {response.reason_phrase})"
+                    )
+                streaming_text = ""
+                async for value in response.aiter_lines():
+                    try:
+                        if bool(value):
+                            streaming_text += value + ("\n" if stream else "")
+                            resp = dict(text=streaming_text)
+                            self.last_response.update(resp)
+                            yield value if raw else resp
+                    except json.decoder.JSONDecodeError:
+                        pass
+            self.conversation.update_chat_history(
+                prompt, await self.get_message(self.last_response)
+            )
+
+        async def for_non_stream():
+            async for _ in for_stream():
+                pass
+            return self.last_response
+
+        return for_stream() if stream else await for_non_stream()
+
+    async def chat(
+        self,
+        prompt: str,
+        stream: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> str | AsyncGenerator:
+        """Generate response `str` asynchronously.
+        Args:
+            prompt (str): Prompt to be send.
+            stream (bool, optional): Flag for streaming response. Defaults to False.
+            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+        Returns:
+            str|AsyncGenerator: Response generated
+        """
+
+        async def for_stream():
+            async_ask = await self.ask(
+                prompt, True, optimizer=optimizer, conversationally=conversationally
+            )
+            async for response in async_ask:
+                yield await self.get_message(response)
+
+        async def for_non_stream():
+            return await self.get_message(
+                await self.ask(
+                    prompt,
+                    False,
+                    optimizer=optimizer,
+                    conversationally=conversationally,
+                )
+            )
+
+        return for_stream() if stream else await for_non_stream()
+
+    async def get_message(self, response: dict) -> str:
+        """Retrieves message only from response
+
+        Args:
+            response (dict): Response generated by `self.ask`
+
+        Returns:
+            str: Message extracted
+        """
+        assert isinstance(response, dict), "Response should be of dict data-type only"
+        return response["text"]
+#------------------------------------------------------phind-------------------------------------------------------------
+class PhindSearch:
+    # default_model = "Phind Model"
+    def __init__(
+        self,
+        is_conversation: bool = True,
+        max_tokens: int = 8000,
+        timeout: int = 30,
+        intro: str = None,
+        filepath: str = None,
+        update_file: bool = True,
+        proxies: dict = {},
+        history_offset: int = 10250,
+        act: str = None,
+        model: str = "Phind Model",
+        quiet: bool = False,
+    ):
+        """Instantiates PHIND
+
+        Args:
+            is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True
+            max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 600.
+            timeout (int, optional): Http request timeout. Defaults to 30.
+            intro (str, optional): Conversation introductory prompt. Defaults to None.
+            filepath (str, optional): Path to file containing conversation history. Defaults to None.
+            update_file (bool, optional): Add new prompts and responses to the file. Defaults to True.
+            proxies (dict, optional): Http request proxies. Defaults to {}.
+            history_offset (int, optional): Limit conversation history to this number of last texts. Defaults to 10250.
+            act (str|int, optional): Awesome prompt key or index. (Used as intro). Defaults to None.
+            model (str, optional): Model name. Defaults to "Phind Model".
+            quiet (bool, optional): Ignore web search-results and yield final response only. Defaults to False.
+        """
+        self.session = requests.Session()
+        self.max_tokens_to_sample = max_tokens
+        self.is_conversation = is_conversation
+        self.chat_endpoint = "https://https.extension.phind.com/agent/"
+        self.stream_chunk_size = 64
+        self.timeout = timeout
+        self.last_response = {}
+        self.model = model
+        self.quiet = quiet
+
+        self.headers = {
+            "Content-Type": "application/json",
+            "User-Agent": "",
+            "Accept": "*/*",
+            "Accept-Encoding": "Identity",
+        }
 
         self.__available_optimizers = (
             method
@@ -1641,27 +3749,19 @@ class OPENGPT:
            dict : {}
         ```json
         {
-            "messages": [
-                {
-                    "content": "Hello there",
-                    "additional_kwargs": {},
-                    "type": "human",
-                    "example": false
-                },
-                {
-                    "content": "Hello! How can I assist you today?",
-                    "additional_kwargs": {
-                    "agent": {
-                        "return_values": {
-                            "output": "Hello! How can I assist you today?"
-                            },
-                        "log": "Hello! How can I assist you today?",
-                        "type": "AgentFinish"
-                    }
-                },
-                "type": "ai",
-                "example": false
-                }]
+            "id": "chatcmpl-r0wujizf2i2xb60mjiwt",
+            "object": "chat.completion.chunk",
+            "created": 1706775384,
+            "model": "trt-llm-phind-model-serving",
+            "choices": [
+                {
+                    "index": 0,
+                    "delta": {
+                        "content": "Hello! How can I assist you with your programming today?"
+                        },
+                    "finish_reason": null
+                }
+            ]
         }
         ```
         """
@@ -1677,22 +3777,15 @@ class OPENGPT:
                 )
 
         self.session.headers.update(self.headers)
-        self.session.headers.update(
-            dict(
-                cookie=f"opengpts_user_id={uuid4().__str__()}",
-            )
-        )
         payload = {
-            "input": [
-                {
-                    "content": conversation_prompt,
-                    "additional_kwargs": {},
-                    "type": "human",
-                    "example": False,
-                },
+            "additional_extension_context": "",
+            "allow_magic_buttons": True,
+            "is_vscode_extension": True,
+            "message_history": [
+                {"content": conversation_prompt, "metadata": {}, "role": "user"}
             ],
-            "assistant_id": self.assistant_id,
-            "thread_id": "",
+            "requested_model": self.model,
+            "user_input": prompt,
         }
 
         def for_stream():
@@ -1704,21 +3797,26 @@ class OPENGPT:
                 or not response.headers.get("Content-Type")
                 == "text/event-stream; charset=utf-8"
             ):
-                raise Exception(
+                raise exceptions.FailedToGenerateResponseError(
                     f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
                 )
-
+            streaming_text = ""
             for value in response.iter_lines(
                 decode_unicode=True,
                 chunk_size=self.stream_chunk_size,
             ):
                 try:
                     modified_value = re.sub("data:", "", value)
-                    resp = json.loads(modified_value)
-                    if len(resp) == 1:
+                    json_modified_value = json.loads(modified_value)
+                    retrieved_text = self.get_message(json_modified_value)
+                    if not retrieved_text:
                         continue
-                    self.last_response.update(resp[1])
-                    yield value if raw else resp[1]
+                    streaming_text += retrieved_text
+                    json_modified_value["choices"][0]["delta"][
+                        "content"
+                    ] = streaming_text
+                    self.last_response.update(json_modified_value)
+                    yield value if raw else json_modified_value
                 except json.decoder.JSONDecodeError:
                     pass
             self.conversation.update_chat_history(
@@ -1777,13 +3875,40 @@ class OPENGPT:
             str: Message extracted
         """
         assert isinstance(response, dict), "Response should be of dict data-type only"
-        return response["content"]
-#------------------------------------------------------PERPLEXITY--------------------------------------------------------  
-class PERPLEXITY:
+        if response.get("type", "") == "metadata":
+            return
+
+        delta: dict = response["choices"][0]["delta"]
+
+        if not delta:
+            return ""
+
+        elif delta.get("function_call"):
+            if self.quiet:
+                return ""
+
+            function_call: dict = delta["function_call"]
+            if function_call.get("name"):
+                return function_call["name"]
+            elif function_call.get("arguments"):
+                return function_call.get("arguments")
+
+        elif delta.get("metadata"):
+            if self.quiet:
+                return ""
+            return yaml.dump(delta["metadata"])
+
+        else:
+            return (
+                response["choices"][0]["delta"].get("content")
+                if response["choices"][0].get("finish_reason") is None
+                else ""
+            )
+class AsyncPhindSearch(AsyncProvider):
     def __init__(
         self,
         is_conversation: bool = True,
-        max_tokens: int = 8000,
+        max_tokens: int = 600,
         timeout: int = 30,
         intro: str = None,
         filepath: str = None,
@@ -1791,9 +3916,10 @@ class PERPLEXITY:
         proxies: dict = {},
         history_offset: int = 10250,
         act: str = None,
+        model: str = "Phind Model",
         quiet: bool = False,
     ):
-        """Instantiates PERPLEXITY
+        """Instantiates PHIND
 
         Args:
             is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True
@@ -1805,16 +3931,25 @@ class PERPLEXITY:
             proxies (dict, optional): Http request proxies. Defaults to {}.
             history_offset (int, optional): Limit conversation history to this number of last texts. Defaults to 10250.
             act (str|int, optional): Awesome prompt key or index. (Used as intro). Defaults to None.
+            model (str, optional): Model name. Defaults to "Phind Model".
             quiet (bool, optional): Ignore web search-results and yield final response only. Defaults to False.
         """
-        logging.getLogger("websocket").setLevel(logging.ERROR)
-        self.session = requests.Session()
         self.max_tokens_to_sample = max_tokens
         self.is_conversation = is_conversation
+        self.chat_endpoint = "https://https.extension.phind.com/agent/"
+        self.stream_chunk_size = 64
+        self.timeout = timeout
         self.last_response = {}
-        self.web_results: dict = {}
+        self.model = model
         self.quiet = quiet
 
+        self.headers = {
+            "Content-Type": "application/json",
+            "User-Agent": "",
+            "Accept": "*/*",
+            "Accept-Encoding": "Identity",
+        }
+
         self.__available_optimizers = (
             method
             for method in dir(Optimizers)
@@ -1831,48 +3966,44 @@ class PERPLEXITY:
             is_conversation, self.max_tokens_to_sample, filepath, update_file
         )
         self.conversation.history_offset = history_offset
+        self.session = httpx.AsyncClient(headers=self.headers, proxies=proxies)
 
-    def ask(
+    async def ask(
         self,
         prompt: str,
         stream: bool = False,
         raw: bool = False,
         optimizer: str = None,
         conversationally: bool = False,
-    ) -> dict:
-        """Chat with AI
+        synchronous_generator=False,
+    ) -> dict | AsyncGenerator:
+        """Asynchronously Chat with AI
 
-                Args:
-                    prompt (str): Prompt to be send.
-                    stream (bool, optional): Flag for streaming response. Defaults to False.
-                    raw (bool, optional): Stream back raw response as received. Defaults to False.
-                    optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
-                    conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
-                Returns:
-                   dict : {}
-                ```json
+        Args:
+            prompt (str): Prompt to be send.
+            stream (bool, optional): Flag for streaming response. Defaults to False.
+            raw (bool, optional): Stream back raw response as received. Defaults to False.
+            optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
+            conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
+        Returns:
+           dict|AsyncGenerator : ai content.
+        ```json
         {
-            "status": "pending",
-            "uuid": "3604dfcc-611f-4b7d-989d-edca2a7233c7",
-            "read_write_token": null,
-            "frontend_context_uuid": "f6d43119-5231-481d-b692-f52e1f52d2c6",
-            "final": false,
-            "backend_uuid": "a6d6ec9e-da69-4841-af74-0de0409267a8",
-            "media_items": [],
-            "widget_data": [],
-            "knowledge_cards": [],
-            "expect_search_results": "false",
-            "mode": "concise",
-            "search_focus": "internet",
-            "gpt4": false,
-            "display_model": "turbo",
-            "attachments": null,
-            "answer": "",
-            "web_results": [],
-            "chunks": [],
-            "extra_web_results": []
+            "id": "chatcmpl-r0wujizf2i2xb60mjiwt",
+            "object": "chat.completion.chunk",
+            "created": 1706775384,
+            "model": "trt-llm-phind-model-serving",
+            "choices": [
+                {
+                    "index": 0,
+                    "delta": {
+                        "content": "Hello! How can I assist you with your programming today?"
+                        },
+                    "finish_reason": null
+                }
+            ]
         }
-                ```
+        ```
         """
         conversation_prompt = self.conversation.gen_complete_prompt(prompt)
         if optimizer:
@@ -1885,30 +4016,70 @@ class PERPLEXITY:
                     f"Optimizer is not one of {self.__available_optimizers}"
                 )
 
-        def for_stream():
-            for response in Perplexity().generate_answer(conversation_prompt):
-                yield json.dumps(response) if raw else response
-                self.last_response.update(response)
+        payload = {
+            "additional_extension_context": "",
+            "allow_magic_buttons": True,
+            "is_vscode_extension": True,
+            "message_history": [
+                {"content": conversation_prompt, "metadata": {}, "role": "user"}
+            ],
+            "requested_model": self.model,
+            "user_input": prompt,
+        }
 
-            self.conversation.update_chat_history(
-                prompt,
-                self.get_message(self.last_response),
-            )
+        async def for_stream():
+            async with self.session.stream(
+                "POST",
+                self.chat_endpoint,
+                json=payload,
+                timeout=self.timeout,
+            ) as response:
+                if (
+                    not response.is_success
+                    or not response.headers.get("Content-Type")
+                    == "text/event-stream; charset=utf-8"
+                ):
+                    raise exceptions.FailedToGenerateResponseError(
+                        f"Failed to generate response - ({response.status_code}, {response.reason_phrase})"
+                    )
+                streaming_text = ""
+                async for value in response.aiter_lines():
+                    try:
+                        modified_value = re.sub("data:", "", value)
+                        json_modified_value = json.loads(modified_value)
+                        retrieved_text = await self.get_message(json_modified_value)
+                        if not retrieved_text:
+                            continue
+                        streaming_text += retrieved_text
+                        json_modified_value["choices"][0]["delta"][
+                            "content"
+                        ] = streaming_text
+                        self.last_response.update(json_modified_value)
+                        yield value if raw else json_modified_value
+                    except json.decoder.JSONDecodeError:
+                        pass
+                self.conversation.update_chat_history(
+                    prompt, await self.get_message(self.last_response)
+                )
 
-        def for_non_stream():
-            for _ in for_stream():
+        async def for_non_stream():
+            async for _ in for_stream():
                 pass
             return self.last_response
 
-        return for_stream() if stream else for_non_stream()
+        return (
+            for_stream()
+            if stream and not synchronous_generator
+            else await for_non_stream()
+        )
 
-    def chat(
+    async def chat(
         self,
         prompt: str,
         stream: bool = False,
         optimizer: str = None,
         conversationally: bool = False,
-    ) -> str:
+    ) -> str | AsyncGenerator:
         """Generate response `str`
         Args:
             prompt (str): Prompt to be send.
@@ -1916,18 +4087,19 @@ class PERPLEXITY:
             optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
             conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
         Returns:
-            str: Response generated
+            str|AsyncGenerator: Response generated
         """
 
-        def for_stream():
-            for response in self.ask(
+        async def for_stream():
+            ask_resp = await self.ask(
                 prompt, True, optimizer=optimizer, conversationally=conversationally
-            ):
-                yield self.get_message(response)
+            )
+            async for response in ask_resp:
+                yield await self.get_message(response)
 
-        def for_non_stream():
-            return self.get_message(
-                self.ask(
+        async def for_non_stream():
+            return await self.get_message(
+                await self.ask(
                     prompt,
                     False,
                     optimizer=optimizer,
@@ -1935,9 +4107,9 @@ class PERPLEXITY:
                 )
             )
 
-        return for_stream() if stream else for_non_stream()
+        return for_stream() if stream else await for_non_stream()
 
-    def get_message(self, response: dict) -> str:
+    async def get_message(self, response: dict) -> str:
         """Retrieves message only from response
 
         Args:
@@ -1947,45 +4119,46 @@ class PERPLEXITY:
             str: Message extracted
         """
         assert isinstance(response, dict), "Response should be of dict data-type only"
-        text_str: str = response.get("answer", "")
+        if response.get("type", "") == "metadata":
+            return
 
-        def update_web_results(web_results: list) -> None:
-            for index, results in enumerate(web_results, start=1):
-                self.web_results[str(index) + ". " + results["name"]] = dict(
-                    url=results.get("url"), snippet=results.get("snippet")
-                )
+        delta: dict = response["choices"][0]["delta"]
 
-        if response.get("text"):
-            # last chunk
-            target: dict[str, Any] = json.loads(response.get("text"))
-            text_str = target.get("answer")
-            web_results: list[dict] = target.get("web_results")
-            self.web_results.clear()
-            update_web_results(web_results)
+        if not delta:
+            return ""
 
-            return (
-                text_str
-                if self.quiet or not self.web_results
-                else text_str + "\n\n# WEB-RESULTS\n\n" + yaml.dump(self.web_results)
-            )
+        elif delta.get("function_call"):
+            if self.quiet:
+                return ""
+
+            function_call: dict = delta["function_call"]
+            if function_call.get("name"):
+                return function_call["name"]
+            elif function_call.get("arguments"):
+                return function_call.get("arguments")
+
+        elif delta.get("metadata"):
+            if self.quiet:
+                return ""
+            return yaml.dump(delta["metadata"])
 
         else:
-            if str(response.get("expect_search_results")).lower() == "true":
-                return (
-                    text_str
-                    if self.quiet
-                    else text_str
-                    + "\n\n# WEB-RESULTS\n\n"
-                    + yaml.dump(response.get("web_results"))
-                )
-            else:
-                return text_str
-#------------------------------------------------------BLACKBOXAI--------------------------------------------------------  
-class BLACKBOXAI:
+            return (
+                response["choices"][0]["delta"].get("content")
+                if response["choices"][0].get("finish_reason") is None
+                else ""
+            )
+#-------------------------------------------------------yep.com--------------------------------------------------------   
+class YEPCHAT(Provider):
     def __init__(
         self,
         is_conversation: bool = True,
-        max_tokens: int = 8000,
+        max_tokens: int = 600,
+        temperature: float = 0.6,
+        presence_penalty: int = 0,
+        frequency_penalty: int = 0,
+        top_p: float = 0.7,
+        model: str ="Mixtral-8x7B-Instruct-v0.1",
         timeout: int = 30,
         intro: str = None,
         filepath: str = None,
@@ -1993,13 +4166,17 @@ class BLACKBOXAI:
         proxies: dict = {},
         history_offset: int = 10250,
         act: str = None,
-        model: str = None,
     ):
-        """Instantiates BLACKBOXAI
+        """Instantiates YEPCHAT
 
         Args:
-            is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True
+            is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True.
             max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 600.
+            temperature (float, optional): Charge of the generated text's randomness. Defaults to 0.6.
+            presence_penalty (int, optional): Chances of topic being repeated. Defaults to 0.
+            frequency_penalty (int, optional): Chances of word being repeated. Defaults to 0.
+            top_p (float, optional): Sampling threshold during inference time. Defaults to 0.7.
+            model (str, optional): LLM model name. Defaults to "gpt-3.5-turbo".
             timeout (int, optional): Http request timeout. Defaults to 30.
             intro (str, optional): Conversation introductory prompt. Defaults to None.
             filepath (str, optional): Path to file containing conversation history. Defaults to None.
@@ -2007,29 +4184,26 @@ class BLACKBOXAI:
             proxies (dict, optional): Http request proxies. Defaults to {}.
             history_offset (int, optional): Limit conversation history to this number of last texts. Defaults to 10250.
             act (str|int, optional): Awesome prompt key or index. (Used as intro). Defaults to None.
-            model (str, optional): Model name. Defaults to "Phind Model".
         """
-        self.session = requests.Session()
-        self.max_tokens_to_sample = max_tokens
         self.is_conversation = is_conversation
-        self.chat_endpoint = "https://www.blackbox.ai/api/chat"
+        self.max_tokens_to_sample = max_tokens
+        self.model = model
+        self.temperature = temperature
+        self.presence_penalty = presence_penalty
+        self.frequency_penalty = frequency_penalty
+        self.top_p = top_p
+        self.chat_endpoint = "https://api.yep.com/v1/chat/completions"
         self.stream_chunk_size = 64
         self.timeout = timeout
         self.last_response = {}
-        self.model = model
-        self.previewToken: str = None
-        self.userId: str = ""
-        self.codeModelMode: bool = True
-        self.id: str = ""
-        self.agentMode: dict = {}
-        self.trendingAgentMode: dict = {}
-        self.isMicMode: bool = False
-
         self.headers = {
-            "Content-Type": "application/json",
-            "User-Agent": "",
             "Accept": "*/*",
-            "Accept-Encoding": "Identity",
+            "Accept-Encoding": "gzip, deflate",
+            "Accept-Language": "en-US,en;q=0.9",
+            "Content-Type": "application/json; charset=utf-8",
+            "Origin": "https://yep.com",
+            "Referer": "https://yep.com/",
+            "User-Agent": "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36",
         }
 
         self.__available_optimizers = (
@@ -2071,7 +4245,20 @@ class BLACKBOXAI:
            dict : {}
         ```json
         {
-           "text" : "print('How may I help you today?')"
+            "id": "cmpl-c61c1c88de4e4ad3a79134775d17ea0c",
+            "object": "chat.completion.chunk",
+            "created": 1713876886,
+            "model": "Mixtral-8x7B-Instruct-v0.1",
+            "choices": [
+                {
+                    "index": 0,
+                    "delta": {
+                        "role": null,
+                        "content": " Sure, I can help with that. Are you looking for information on how to start coding, or do you need help with a specific coding problem? We can discuss various programming languages like Python, JavaScript, Java, C++, or others. Please provide more details so I can assist you better."
+                        },
+                    "finish_reason": null
+                }
+            ]
         }
         ```
         """
@@ -2085,47 +4272,41 @@ class BLACKBOXAI:
                 raise Exception(
                     f"Optimizer is not one of {self.__available_optimizers}"
                 )
-
         self.session.headers.update(self.headers)
         payload = {
-            "messages": [
-                # json.loads(prev_messages),
-                {"content": conversation_prompt, "role": "user"}
-            ],
-            "id": self.id,
-            "previewToken": self.previewToken,
-            "userId": self.userId,
-            "codeModelMode": self.codeModelMode,
-            "agentMode": self.agentMode,
-            "trendingAgentMode": self.trendingAgentMode,
-            "isMicMode": self.isMicMode,
+            "stream": True,
+            "max_tokens": 1280,
+            "top_p": self.top_p,
+            "temperature": self.temperature,
+            "messages": [{"content": conversation_prompt, "role": "user"}],
+            "model": self.model,
         }
 
         def for_stream():
             response = self.session.post(
                 self.chat_endpoint, json=payload, stream=True, timeout=self.timeout
             )
-            if (
-                not response.ok
-                or not response.headers.get("Content-Type")
-                == "text/plain; charset=utf-8"
-            ):
-                raise Exception(
+            if not response.ok:
+                raise exceptions.FailedToGenerateResponseError(
                     f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
                 )
-            streaming_text = ""
+
+            message_load = ""
             for value in response.iter_lines(
                 decode_unicode=True,
+                delimiter="" if raw else "data:",
                 chunk_size=self.stream_chunk_size,
-                delimiter="\n",
             ):
                 try:
-                    if bool(value):
-                        streaming_text += value + ("\n" if stream else "")
-
-                        resp = dict(text=streaming_text)
+                    resp = json.loads(value)
+                    incomplete_message = self.get_message(resp)
+                    if incomplete_message:
+                        message_load += incomplete_message
+                        resp["choices"][0]["delta"]["content"] = message_load
                         self.last_response.update(resp)
                         yield value if raw else resp
+                    elif raw:
+                        yield value
                 except json.decoder.JSONDecodeError:
                     pass
             self.conversation.update_chat_history(
@@ -2184,21 +4365,22 @@ class BLACKBOXAI:
             str: Message extracted
         """
         assert isinstance(response, dict), "Response should be of dict data-type only"
-        return response["text"]
-    @staticmethod
-    def chat_cli(prompt):
-        """Sends a request to the BLACKBOXAI API and processes the response."""
-        blackbox_ai = BLACKBOXAI()  # Initialize a BLACKBOXAI instance
-        response = blackbox_ai.ask(prompt)  # Perform a chat with the given prompt
-        processed_response = blackbox_ai.get_message(response)  # Process the response
-        print(processed_response)
-#------------------------------------------------------phind-------------------------------------------------------------
-class PhindSearch:
-    # default_model = "Phind Model"
+        try:
+            if response["choices"][0].get("delta"):
+                return response["choices"][0]["delta"]["content"]
+            return response["choices"][0]["message"]["content"]
+        except KeyError:
+            return ""
+class AsyncYEPCHAT(AsyncProvider):
     def __init__(
         self,
         is_conversation: bool = True,
-        max_tokens: int = 8000,
+        max_tokens: int = 600,
+        temperature: float = 0.6,
+        presence_penalty: int = 0,
+        frequency_penalty: int = 0,
+        top_p: float = 0.7,
+        model: str = "Mixtral-8x7B-Instruct-v0.1",
         timeout: int = 30,
         intro: str = None,
         filepath: str = None,
@@ -2206,14 +4388,17 @@ class PhindSearch:
         proxies: dict = {},
         history_offset: int = 10250,
         act: str = None,
-        model: str = "Phind Model",
-        quiet: bool = False,
     ):
-        """Instantiates PHIND
+        """Instantiates YEPCHAT
 
         Args:
-            is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True
+            is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True.
             max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 600.
+            temperature (float, optional): Charge of the generated text's randomness. Defaults to 0.6.
+            presence_penalty (int, optional): Chances of topic being repeated. Defaults to 0.
+            frequency_penalty (int, optional): Chances of word being repeated. Defaults to 0.
+            top_p (float, optional): Sampling threshold during inference time. Defaults to 0.7.
+            model (str, optional): LLM model name. Defaults to "gpt-3.5-turbo".
             timeout (int, optional): Http request timeout. Defaults to 30.
             intro (str, optional): Conversation introductory prompt. Defaults to None.
             filepath (str, optional): Path to file containing conversation history. Defaults to None.
@@ -2221,24 +4406,26 @@ class PhindSearch:
             proxies (dict, optional): Http request proxies. Defaults to {}.
             history_offset (int, optional): Limit conversation history to this number of last texts. Defaults to 10250.
             act (str|int, optional): Awesome prompt key or index. (Used as intro). Defaults to None.
-            model (str, optional): Model name. Defaults to "Phind Model".
-            quiet (bool, optional): Ignore web search-results and yield final response only. Defaults to False.
         """
-        self.session = requests.Session()
-        self.max_tokens_to_sample = max_tokens
         self.is_conversation = is_conversation
-        self.chat_endpoint = "https://https.extension.phind.com/agent/"
+        self.max_tokens_to_sample = max_tokens
+        self.model = model
+        self.temperature = temperature
+        self.presence_penalty = presence_penalty
+        self.frequency_penalty = frequency_penalty
+        self.top_p = top_p
+        self.chat_endpoint = "https://api.yep.com/v1/chat/completions"
         self.stream_chunk_size = 64
         self.timeout = timeout
         self.last_response = {}
-        self.model = model
-        self.quiet = quiet
-
         self.headers = {
-            "Content-Type": "application/json",
-            "User-Agent": "",
             "Accept": "*/*",
-            "Accept-Encoding": "Identity",
+            "Accept-Encoding": "gzip, deflate",
+            "Accept-Language": "en-US,en;q=0.9",
+            "Content-Type": "application/json; charset=utf-8",
+            "Origin": "https://yep.com",
+            "Referer": "https://yep.com/",
+            "User-Agent": "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36",
         }
 
         self.__available_optimizers = (
@@ -2246,7 +4433,6 @@ class PhindSearch:
             for method in dir(Optimizers)
             if callable(getattr(Optimizers, method)) and not method.startswith("__")
         )
-        self.session.headers.update(self.headers)
         Conversation.intro = (
             AwesomePrompts().get_act(
                 act, raise_not_found=True, default=None, case_insensitive=True
@@ -2258,9 +4444,12 @@ class PhindSearch:
             is_conversation, self.max_tokens_to_sample, filepath, update_file
         )
         self.conversation.history_offset = history_offset
-        self.session.proxies = proxies
+        self.session = httpx.AsyncClient(
+            headers=self.headers,
+            proxies=proxies,
+        )
 
-    def ask(
+    async def ask(
         self,
         prompt: str,
         stream: bool = False,
@@ -2268,7 +4457,7 @@ class PhindSearch:
         optimizer: str = None,
         conversationally: bool = False,
     ) -> dict:
-        """Chat with AI
+        """Chat with AI asynchronously.
 
         Args:
             prompt (str): Prompt to be send.
@@ -2280,15 +4469,16 @@ class PhindSearch:
            dict : {}
         ```json
         {
-            "id": "chatcmpl-r0wujizf2i2xb60mjiwt",
+            "id": "cmpl-c61c1c88de4e4ad3a79134775d17ea0c",
             "object": "chat.completion.chunk",
-            "created": 1706775384,
-            "model": "trt-llm-phind-model-serving",
+            "created": 1713876886,
+            "model": "Mixtral-8x7B-Instruct-v0.1",
             "choices": [
                 {
                     "index": 0,
                     "delta": {
-                        "content": "Hello! How can I assist you with your programming today?"
+                        "role": null,
+                        "content": " Sure, I can help with that. Are you looking for information on how to start coding, or do you need help with a specific coding problem? We can discuss various programming languages like Python, JavaScript, Java, C++, or others. Please provide more details so I can assist you better."
                         },
                     "finish_reason": null
                 }
@@ -2306,69 +4496,58 @@ class PhindSearch:
                 raise Exception(
                     f"Optimizer is not one of {self.__available_optimizers}"
                 )
-
-        self.session.headers.update(self.headers)
         payload = {
-            "additional_extension_context": "",
-            "allow_magic_buttons": True,
-            "is_vscode_extension": True,
-            "message_history": [
-                {"content": conversation_prompt, "metadata": {}, "role": "user"}
-            ],
-            "requested_model": self.model,
-            "user_input": prompt,
+            "stream": True,
+            "max_tokens": 1280,
+            "top_p": self.top_p,
+            "temperature": self.temperature,
+            "messages": [{"content": conversation_prompt, "role": "user"}],
+            "model": self.model,
         }
 
-        def for_stream():
-            response = self.session.post(
-                self.chat_endpoint, json=payload, stream=True, timeout=self.timeout
-            )
-            if (
-                not response.ok
-                or not response.headers.get("Content-Type")
-                == "text/event-stream; charset=utf-8"
-            ):
-                raise Exception(
-                    f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
-                )
-            streaming_text = ""
-            for value in response.iter_lines(
-                decode_unicode=True,
-                chunk_size=self.stream_chunk_size,
-            ):
-                try:
-                    modified_value = re.sub("data:", "", value)
-                    json_modified_value = json.loads(modified_value)
-                    retrieved_text = self.get_message(json_modified_value)
-                    if not retrieved_text:
-                        continue
-                    streaming_text += retrieved_text
-                    json_modified_value["choices"][0]["delta"][
-                        "content"
-                    ] = streaming_text
-                    self.last_response.update(json_modified_value)
-                    yield value if raw else json_modified_value
-                except json.decoder.JSONDecodeError:
-                    pass
+        async def for_stream():
+            async with self.session.stream(
+                "POST", self.chat_endpoint, json=payload, timeout=self.timeout
+            ) as response:
+                if not response.is_success:
+                    raise exceptions.FailedToGenerateResponseError(
+                        f"Failed to generate response - ({response.status_code}, {response.reason_phrase}) - {response.text}"
+                    )
+
+                message_load = ""
+                async for value in response.aiter_lines():
+                    try:
+                        resp = sanitize_stream(value)
+                        incomplete_message = await self.get_message(resp)
+                        if incomplete_message:
+                            message_load += incomplete_message
+                            resp["choices"][0]["delta"]["content"] = message_load
+                            self.last_response.update(resp)
+                            yield value if raw else resp
+                        elif raw:
+                            yield value
+                    except json.decoder.JSONDecodeError:
+                        pass
+
             self.conversation.update_chat_history(
-                prompt, self.get_message(self.last_response)
+                prompt, await self.get_message(self.last_response)
             )
 
-        def for_non_stream():
-            for _ in for_stream():
+        async def for_non_stream():
+            async for _ in for_stream():
                 pass
             return self.last_response
 
-        return for_stream() if stream else for_non_stream()
+        return for_stream() if stream else await for_non_stream()
 
-    def chat(
+    async def chat(
         self,
         prompt: str,
         stream: bool = False,
         optimizer: str = None,
         conversationally: bool = False,
     ) -> str:
-        """Generate response `str`
+        """Generate response `str` asynchronously.
         Args:
             prompt (str): Prompt to be send.
             stream (bool, optional): Flag for streaming response. Defaults to False.
@@ -2378,15 +4557,17 @@ class PhindSearch:
             str: Response generated
         """
 
-        def for_stream():
-            for response in self.ask(
+        async def for_stream():
+            async_ask = await self.ask(
                 prompt, True, optimizer=optimizer, conversationally=conversationally
-            ):
-                yield self.get_message(response)
+            )
 
-        def for_non_stream():
-            return self.get_message(
-                self.ask(
+            async for response in async_ask:
+                yield await self.get_message(response)
+
+        async def for_non_stream():
+            return await self.get_message(
+                await self.ask(
                     prompt,
                     False,
                     optimizer=optimizer,
@@ -2394,9 +4575,9 @@ class PhindSearch:
                 )
             )
 
-        return for_stream() if stream else for_non_stream()
+        return for_stream() if stream else await for_non_stream()
 
-    def get_message(self, response: dict) -> str:
+    async def get_message(self, response: dict) -> str:
         """Retrieves message only from response
 
         Args:
@@ -2406,99 +4587,12 @@ class PhindSearch:
             str: Message extracted
         """
         assert isinstance(response, dict), "Response should be of dict data-type only"
-        if response.get("type", "") == "metadata":
-            return
-
-        delta: dict = response["choices"][0]["delta"]
-
-        if not delta:
+        try:
+            if response["choices"][0].get("delta"):
+                return response["choices"][0]["delta"]["content"]
+            return response["choices"][0]["message"]["content"]
+        except KeyError:
             return ""
-
-        elif delta.get("function_call"):
-            if self.quiet:
-                return ""
-
-            function_call: dict = delta["function_call"]
-            if function_call.get("name"):
-                return function_call["name"]
-            elif function_call.get("arguments"):
-                return function_call.get("arguments")
-
-        elif delta.get("metadata"):
-            if self.quiet:
-                return ""
-            return yaml.dump(delta["metadata"])
-
-        else:
-            return (
-                response["choices"][0]["delta"].get("content")
-                if response["choices"][0].get("finish_reason") is None
-                else ""
-            )
-    @staticmethod
-    def chat_cli(prompt):
-        """Sends a request to the Phind API and processes the response."""
-        phind_search = PhindSearch()  # Initialize a PhindSearch instance
-        response = phind_search.ask(prompt)  # Perform a search with the given prompt
-        processed_response = phind_search.get_message(response)  # Process the response
-        print(processed_response)
-#-------------------------------------------------------yep.com--------------------------------------------------------   
-class YepChat:
-    def __init__(self, message="hello"):
-        self.url = "https://api.yep.com/v1/chat/completions"
-        self.headers = {
-            "Accept": "*/*",
-            "Accept-Encoding": "gzip, deflate, br, zstd",
-            "Accept-Language": "en-US,en;q=0.9",
-            "Cache-Control": "max-age=0",
-            "Content-Type": "application/json; charset=utf-8",
-            "Origin": "https://yep.com",
-            "Referer": "https://yep.com/",
-            "Sec-Ch-Ua": '"Chromium";v="122", "Not(A:Brand";v="24", "Google Chrome";v="122"',
-            "Sec-Ch-Ua-Mobile": "?0",
-            "Sec-Ch-Ua-Platform": '"Windows"',
-            "Sec-Fetch-Dest": "empty",
-            "Sec-Fetch-Mode": "cors",
-            "Sec-Fetch-Site": "same-site",
-            "User-Agent": "Mozilla/5.0 (Windows NT   10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/122.0.0.0.0.0 Safari/537.36"
-        }
-        self.payload = {
-            "stream": True,
-            "max_tokens":   1280,
-            "top_p":   0.7,
-            "temperature":   0.6,
-            "messages": [{
-                "content": message,
-                "role": "user"
-            }],
-            "model": "Mixtral-8x7B-Instruct-v0.1"
-        }
-
-    def send_request(self):
-        response = requests.post(self.url, headers=self.headers, data=json.dumps(self.payload), stream=True)
-        print(response.status_code)
-        return response
-
-    def process_response(self, response):
-        myset = ""
-        for line in response.iter_lines():
-            if line:
-                myline = line.decode('utf-8').removeprefix("data: ").replace(" null", "False")
-                try:
-                    myval = eval(myline)
-                    if "choices" in myval and "delta" in myval["choices"][0] and "content" in myval["choices"][0]["delta"]:
-                        myset += myval["choices"][0]["delta"]["content"]
-                except:
-                    continue
-        return myset
-
-    @staticmethod
-    def chat_cli(message):
-        """Sends a request to the Yep API and processes the response."""
-        yep_chat = YepChat(message=message)
-        response = yep_chat.send_request()
-        processed_response = yep_chat.process_response(response)
-        print(processed_response)
 #-------------------------------------------------------youchat--------------------------------------------------------   
 class youChat:
     """
@@ -2850,88 +4944,3 @@ class Pollinations:
             image.show()
         except Exception as e:
             print(f"An error occurred: {e}")
-
-@click.group()
-def cli():
-    """Webscout AI command-line interface."""
-    pass
-
-@cli.command()
-@click.option('--prompt', prompt='Enter your search prompt', help='The prompt to send.')
-def phindsearch(prompt):
-    """Perform a search with the given prompt using PhindSearch."""
-    phind_search = PhindSearch() # Initialize a PhindSearch instance
-    response = phind_search.ask(prompt) # Perform a search with the given prompt
-    processed_response = phind_search.get_message(response) # Process the response
-    print(processed_response)
-
-@cli.command()
-@click.option('--message', prompt='Enter your message', help='The message to send.')
-def yepchat(message):
-    YepChat.chat_cli(message)
-
-@cli.command()
-@click.option('--prompt', prompt='Enter your prompt', help='The prompt to generate a completion from.')
-def youchat(prompt):
-    youChat.chat_cli(prompt)
-
-
-@cli.command()
-@click.option('--prompt', prompt='Enter your prompt', help='The prompt for generating the image.')
-def prodia(prompt):
-    """Generate an image based on the provided prompt."""
-    Prodia.prodia_cli(prompt)
-
-@cli.command()
-@click.option('--prompt', prompt='Enter your prompt', help='The prompt for generating the image.')
-def pollinations(prompt):
-    """Generate an image based on the provided prompt."""
-    Pollinations.pollinations_cli(prompt)
-    
-@cli.command()
-@click.option('--prompt', prompt='Enter your prompt', help='The prompt to send.')
-def blackboxai(prompt):
-    """Chat with BLACKBOXAI using the provided prompt."""
-    BLACKBOXAI.chat_cli(prompt)
-    
-@cli.command()
-@click.option('--prompt', prompt='Enter your prompt', help='The prompt to send.')
-@click.option('--stream', is_flag=True, help='Flag for streaming response.')
-@click.option('--raw', is_flag=True, help='Stream back raw response as received.')
-@click.option('--optimizer', type=str, help='Prompt optimizer name.')
-@click.option('--conversationally', is_flag=True, help='Chat conversationally when using optimizer.')
-def perplexity(prompt, stream, raw, optimizer, conversationally):
-    """Chat with PERPLEXITY using the provided prompt."""
-    perplexity_instance = PERPLEXITY() # Initialize a PERPLEXITY instance
-    response = perplexity_instance.ask(prompt, stream, raw, optimizer, conversationally)
-    processed_response = perplexity_instance.get_message(response) # Process the response
-    print(processed_response)
-   
-@cli.command()
-@click.option('--prompt', prompt='Enter your search prompt', help='The prompt to send.')
-@click.option('--stream', is_flag=True, help='Flag for streaming response.')
-def opengpt(prompt, stream):
-    """Chat with OPENGPT using the provided prompt."""
-    opengpt = OPENGPT(is_conversation=True, max_tokens=8000, timeout=30)
-    if stream:
-        for response in opengpt.chat(prompt, stream=True):
-            print(response)
-    else:
-        response_str = opengpt.chat(prompt)
-        print(response_str)
-        
-@cli.command()
-@click.option('--prompt', prompt='Enter your prompt', help='The prompt to send.')
-@click.option('--stream', is_flag=True, help='Flag for streaming response.')
-@click.option('--raw', is_flag=True, help='Stream back raw response as received.')
-@click.option('--optimizer', type=str, help='Prompt optimizer name.')
-@click.option('--conversationally', is_flag=True, help='Chat conversationally when using optimizer.')
-def koboldai_cli(prompt, stream, raw, optimizer, conversationally):
-    """Chat with KOBOLDAI using the provided prompt."""
-    koboldai_instance = KOBOLDAI() # Initialize a KOBOLDAI instance
-    response = koboldai_instance.ask(prompt, stream, raw, optimizer, conversationally)
-    processed_response = koboldai_instance.get_message(response) # Process the response
-    print(processed_response)
-    
-if __name__ == '__main__':
-    cli()