webscout 8.2.2__py3-none-any.whl → 2026.1.19__py3-none-any.whl

webscout/Provider/OPENAI/chatgpt.py

@@ -1,18 +1,30 @@
+import base64
+import hashlib
+import json
+import random
 import time
 import uuid
+from datetime import datetime, timezone
+from typing import Any, Dict, Generator, List, Optional, Union, cast
+
 import requests
-import json
-import random
-import base64
-import hashlib
-from datetime import datetime, timedelta
-from typing import List, Dict, Optional, Union, Generator, Any
 
 # Import base classes and utility structures
-from .base import OpenAICompatibleProvider, BaseChat, BaseCompletions
-from .utils import (
-    ChatCompletionChunk, ChatCompletion, Choice, ChoiceDelta,
-    ChatCompletionMessage, CompletionUsage
+from webscout.Provider.OPENAI.base import (
+    BaseChat,
+    BaseCompletions,
+    OpenAICompatibleProvider,
+    SimpleModelList,
+    Tool,
+)
+from webscout.Provider.OPENAI.utils import (
+    ChatCompletion,
+    ChatCompletionChunk,
+    ChatCompletionMessage,
+    Choice,
+    ChoiceDelta,
+    CompletionUsage,
+    count_tokens,
 )
 
 # ANSI escape codes for formatting
@@ -20,14 +32,50 @@ BOLD = "\033[1m"
 RED = "\033[91m"
 RESET = "\033[0m"
 
+
 class ChatGPTReversed:
+    AVAILABLE_MODELS = [
+        "auto",
+        "gpt-5-1",
+        "gpt-5-1-instant",
+        "gpt-5-1-thinking",
+        "gpt-5",
+        "gpt-5-instant",
+        "gpt-5-thinking",
+        "gpt-4",
+        "gpt-4.1",
+        "gpt-4-1",
+        "gpt-4.1-mini",
+        "gpt-4-1-mini",
+        "gpt-4.5",
+        "gpt-4-5",
+        "gpt-4o",
+        "gpt-4o-mini",
+        "o1",
+        "o1-mini",
+        "o3-mini",
+        "o3-mini-high",
+        "o4-mini",
+        "o4-mini-high",
+    ]
     csrf_token = None
     initialized = False
-    AVAILABLE_MODELS = ["auto", "gpt-4o-mini", "gpt-4o", "o4-mini"]
+
+    _instance = None
+
+    def __new__(cls, model="auto"):
+        if cls._instance is None:
+            cls._instance = super(ChatGPTReversed, cls).__new__(cls)
+            cls._instance.initialized = False
+        return cls._instance
 
     def __init__(self, model="auto"):
-        if ChatGPTReversed.initialized:
-            raise Exception("ChatGPTReversed has already been initialized.")
+        if self.initialized:
+            # Already initialized, just update model if needed
+            if model not in self.AVAILABLE_MODELS:
+                raise ValueError(f"Invalid model: {model}. Choose from: {self.AVAILABLE_MODELS}")
+            self.model = model
+            return
 
         if model not in self.AVAILABLE_MODELS:
             raise ValueError(f"Invalid model: {model}. Choose from: {self.AVAILABLE_MODELS}")
@@ -56,7 +104,7 @@ class ChatGPTReversed:
             "agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/132.0.0.0 Safari/537.36",
             "platform": "Windows",
             "mobile": "?0",
-            "ua": 'Not A(Brand";v="8", "Chromium";v="132", "Google Chrome";v="132'
+            "ua": 'Not A(Brand";v="8", "Chromium";v="132", "Google Chrome";v="132',
         }
 
         ip = self.random_ip()
@@ -80,58 +128,108 @@ class ChatGPTReversed:
         }
 
         if spoof_address:
-            headers.update({
-                "X-Forwarded-For": ip,
-                "X-Originating-IP": ip,
-                "X-Remote-IP": ip,
-                "X-Remote-Addr": ip,
-                "X-Host": ip,
-                "X-Forwarded-Host": ip
-            })
+            headers.update(
+                {
+                    "X-Forwarded-For": ip,
+                    "X-Originating-IP": ip,
+                    "X-Remote-IP": ip,
+                    "X-Remote-Addr": ip,
+                    "X-Host": ip,
+                    "X-Forwarded-Host": ip,
+                }
+            )
 
         return headers
 
-    def solve_sentinel_challenge(self, seed, difficulty):
-        """Solve the sentinel challenge for authentication."""
-        cores = [8, 12, 16, 24]
-        screens = [3000, 4000, 6000]
-        core = random.choice(cores)
-        screen = random.choice(screens)
-
-        # Adjust time to match expected timezone
-        now = datetime.now() - timedelta(hours=8)
-        parse_time = now.strftime("%a, %d %b %Y %H:%M:%S GMT+0100 (Central European Time)")
+    def generate_proof_token(self, seed: str, difficulty: str, user_agent: Optional[str] = None):
+        """
+        Improved proof-of-work implementation based on gpt4free/g4f/Provider/openai/proofofwork.py
 
-        config = [core + screen, parse_time, 4294705152, 0,
-                 "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/132.0.0.0 Safari/537.36"]
+        Args:
+            seed: The seed string for the challenge
+            difficulty: The difficulty hex string
+            user_agent: Optional user agent string
 
-        diff_len = len(difficulty) // 2
+        Returns:
+            The proof token starting with 'gAAAAAB'
+        """
+        if user_agent is None:
+            user_agent = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/132.0.0.0 Safari/537.36"
+
+        screen = random.choice([3008, 4010, 6000]) * random.choice([1, 2, 4])
+
+        # Get current UTC time
+        now_utc = datetime.now(timezone.utc)
+        parse_time = now_utc.strftime("%a, %d %b %Y %H:%M:%S GMT")
+
+        proof_token = [
+            screen,
+            parse_time,
+            None,
+            0,
+            user_agent,
+            "https://tcr9i.chat.openai.com/v2/35536E1E-65B4-4D96-9D97-6ADB7EFF8147/api.js",
+            "dpl=1440a687921de39ff5ee56b92807faaadce73f13",
+            "en",
+            "en-US",
+            None,
+            "plugins−[object PluginArray]",
+            random.choice(
+                [
+                    "_reactListeningcfilawjnerp",
+                    "_reactListening9ne2dfo1i47",
+                    "_reactListening410nzwhan2a",
+                ]
+            ),
+            random.choice(["alert", "ontransitionend", "onprogress"]),
+        ]
 
+        diff_len = len(difficulty)
         for i in range(100000):
-            config[3] = i
-            json_data = json.dumps(config)
+            proof_token[3] = i
+            json_data = json.dumps(proof_token)
             base = base64.b64encode(json_data.encode()).decode()
-            hash_value = hashlib.sha3_512((seed + base).encode()).hexdigest()
+            hash_value = hashlib.sha3_512((seed + base).encode()).digest()
 
-            if hash_value[:diff_len] <= difficulty:
-                result = "gAAAAAB" + base
-                return result
+            if hash_value.hex()[:diff_len] <= difficulty:
+                return "gAAAAAB" + base
 
         # Fallback
-        fallback_base = base64.b64encode(seed.encode()).decode()
+        fallback_base = base64.b64encode(f'"{seed}"'.encode()).decode()
         return "gAAAAABwQ8Lk5FbGpA2NcR9dShT6gYjU7VxZ4D" + fallback_base
 
+    def solve_sentinel_challenge(self, seed, difficulty):
+        """Solve the sentinel challenge for authentication using improved algorithm."""
+        return self.generate_proof_token(seed, difficulty)
+
     def generate_fake_sentinel_token(self):
         """Generate a fake sentinel token for initial authentication."""
         prefix = "gAAAAAC"
+
+        # More realistic screen sizes
+        screen = random.choice([3008, 4010, 6000]) * random.choice([1, 2, 4])
+
+        # Get current UTC time
+        now_utc = datetime.now(timezone.utc)
+        parse_time = now_utc.strftime("%a, %d %b %Y %H:%M:%S GMT")
+
         config = [
-            random.randint(3000, 6000),
-            datetime.now().strftime("%a, %d %b %Y %H:%M:%S GMT+0100 (Central European Time)"),
-            4294705152, 0,
+            screen,
+            parse_time,
+            4294705152,
+            0,
             "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/132.0.0.0 Safari/537.36",
-            "de", "de", 401, "mediaSession", "location", "scrollX",
+            "de",
+            "de",
+            401,
+            "mediaSession",
+            "location",
+            "scrollX",
             self.random_float(1000, 5000),
-            str(uuid.uuid4()), "", 12, int(time.time() * 1000)
+            str(uuid.uuid4()),
+            "",
+            12,
+            int(time.time() * 1000),
         ]
 
         base64_str = base64.b64encode(json.dumps(config).encode()).decode()
@@ -152,14 +250,16 @@ class ChatGPTReversed:
             try:
                 if part.startswith("data: "):
                     json_data = json.loads(part[6:])
-                    if (json_data.get("message") and
-                        json_data["message"].get("status") == "finished_successfully" and
-                        json_data["message"].get("metadata", {}).get("is_complete")):
+                    if (
+                        json_data.get("message")
+                        and json_data["message"].get("status") == "finished_successfully"
+                        and json_data["message"].get("metadata", {}).get("is_complete")
+                    ):
                         return json_data["message"]["content"]["parts"][0]
-            except:
+            except Exception:
                 pass
 
-        return input_text # Return raw text if parsing fails or no complete message found
+        return input_text  # Return raw text if parsing fails or no complete message found
 
     def rotate_session_data(self):
         """Rotate session data to maintain fresh authentication."""
@@ -169,11 +269,7 @@ class ChatGPTReversed:
 
         ChatGPTReversed.csrf_token = csrf_token
 
-        return {
-            "uuid": uuid_val,
-            "csrf": csrf_token,
-            "sentinel": sentinel_token
-        }
+        return {"uuid": uuid_val, "csrf": csrf_token, "sentinel": sentinel_token}
 
     def get_csrf_token(self, uuid_val):
         """Get CSRF token for authentication."""
@@ -181,15 +277,10 @@ class ChatGPTReversed:
             return ChatGPTReversed.csrf_token
 
         headers = self.simulate_bypass_headers(
-            accept="application/json",
-            spoof_address=True,
-            pre_oai_uuid=uuid_val
+            accept="application/json", spoof_address=True, pre_oai_uuid=uuid_val
         )
 
-        response = requests.get(
-            "https://chatgpt.com/api/auth/csrf",
-            headers=headers
-        )
+        response = requests.get("https://chatgpt.com/api/auth/csrf", headers=headers)
 
         data = response.json()
         if "csrfToken" not in data:
@@ -200,9 +291,7 @@ class ChatGPTReversed:
     def get_sentinel_token(self, uuid_val, csrf):
         """Get sentinel token for authentication."""
         headers = self.simulate_bypass_headers(
-            accept="application/json",
-            spoof_address=True,
-            pre_oai_uuid=uuid_val
+            accept="application/json", spoof_address=True, pre_oai_uuid=uuid_val
         )
 
         test = self.generate_fake_sentinel_token()
@@ -212,8 +301,8 @@ class ChatGPTReversed:
             json={"p": test},
             headers={
                 **headers,
-                "Cookie": f"__Host-next-auth.csrf-token={csrf}; oai-did={uuid_val}; oai-nav-state=1;"
-            }
+                "Cookie": f"__Host-next-auth.csrf-token={csrf}; oai-did={uuid_val}; oai-nav-state=1;",
+            },
         )
 
         data = response.json()
@@ -230,15 +319,10 @@ class ChatGPTReversed:
             raise Exception("Failed to fetch required oai-sc token")
 
         challenge_token = self.solve_sentinel_challenge(
-            data["proofofwork"]["seed"],
-            data["proofofwork"]["difficulty"]
+            data["proofofwork"]["seed"], data["proofofwork"]["difficulty"]
         )
 
-        return {
-            "token": data["token"],
-            "proof": challenge_token,
-            "oaiSc": oai_sc
-        }
+        return {"token": data["token"], "proof": challenge_token, "oaiSc": oai_sc}
 
     def complete(self, message, model=None):
         """Complete a message using ChatGPT.
@@ -252,42 +336,45 @@ class ChatGPTReversed:
             The complete response as a string.
         """
         if not ChatGPTReversed.initialized:
-            raise Exception("ChatGPTReversed has not been initialized. Please initialize the instance before calling this method.")
+            raise Exception(
+                "ChatGPTReversed has not been initialized. Please initialize the instance before calling this method."
+            )
 
         # Use the provided model or fall back to the instance model
         selected_model = model if model else self.model
 
         # Validate the model
         if selected_model not in self.AVAILABLE_MODELS:
-            raise ValueError(f"Invalid model: {selected_model}. Choose from: {self.AVAILABLE_MODELS}")
+            raise ValueError(
+                f"Invalid model: {selected_model}. Choose from: {self.AVAILABLE_MODELS}"
+            )
 
         session_data = self.rotate_session_data()
 
         headers = self.simulate_bypass_headers(
-            accept="plain/text", # Changed accept header as we expect full response now
+            accept="plain/text",  # Changed accept header as we expect full response now
             spoof_address=True,
-            pre_oai_uuid=session_data["uuid"]
+            pre_oai_uuid=session_data["uuid"],
         )
 
-        headers.update({
-            "Cookie": f"__Host-next-auth.csrf-token={session_data['csrf']}; oai-did={session_data['uuid']}; oai-nav-state=1; oai-sc={session_data['sentinel']['oaiSc']};",
-            "openai-sentinel-chat-requirements-token": session_data["sentinel"]["token"],
-            "openai-sentinel-proof-token": session_data["sentinel"]["proof"]
-        })
+        headers.update(
+            {
+                "Cookie": f"__Host-next-auth.csrf-token={session_data['csrf']}; oai-did={session_data['uuid']}; oai-nav-state=1; oai-sc={session_data['sentinel']['oaiSc']};",
+                "openai-sentinel-chat-requirements-token": session_data["sentinel"]["token"],
+                "openai-sentinel-proof-token": session_data["sentinel"]["proof"],
+            }
+        )
 
         payload = {
             "action": "next",
-            "messages": [{
-                "id": self.random_uuid(),
-                "author": {
-                    "role": "user"
-                },
-                "content": {
-                    "content_type": "text",
-                    "parts": [message]
-                },
-                "metadata": {}
-            }],
+            "messages": [
+                {
+                    "id": self.random_uuid(),
+                    "author": {"role": "user"},
+                    "content": {"content_type": "text", "parts": [message]},
+                    "metadata": {},
+                }
+            ],
             "parent_message_id": self.random_uuid(),
             "model": selected_model,  # Use the selected model
             "timezone_offset_min": -120,
@@ -295,7 +382,7 @@ class ChatGPTReversed:
             "history_and_training_disabled": False,
             "conversation_mode": {
                 "kind": "primary_assistant",
-                "plugin_ids": None # Ensure web search is not used
+                "plugin_ids": None,  # Ensure web search is not used
             },
             "force_paragen": False,
             "force_paragen_model_slug": "",
@@ -303,13 +390,11 @@ class ChatGPTReversed:
             "force_rate_limit": False,
             "reset_rate_limits": False,
             "websocket_request_id": self.random_uuid(),
-            "force_use_sse": True # Keep SSE for receiving the full response
+            "force_use_sse": True,  # Keep SSE for receiving the full response
         }
 
         response = requests.post(
-            "https://chatgpt.com/backend-anon/conversation",
-            json=payload,
-            headers=headers
+            "https://chatgpt.com/backend-anon/conversation", json=payload, headers=headers
         )
 
         if response.status_code != 200:
@@ -319,7 +404,7 @@ class ChatGPTReversed:
 
 
 class Completions(BaseCompletions):
-    def __init__(self, client: 'ChatGPT'):
+    def __init__(self, client: "ChatGPT"):
         self._client = client
         self._chatgpt_reversed = None
 
@@ -332,7 +417,13 @@ class Completions(BaseCompletions):
         stream: bool = False,
         temperature: Optional[float] = None,
         top_p: Optional[float] = None,
-        **kwargs: Any
+        tools: Optional[List[Union[Tool, Dict[str, Any]]]] = None,  # Support for tool definitions
+        tool_choice: Optional[
+            Union[str, Dict[str, Any]]
+        ] = None,  # Support for tool_choice parameter
+        timeout: Optional[int] = None,
+        proxies: Optional[dict] = None,
+        **kwargs: Any,
     ) -> Union[ChatCompletion, Generator[ChatCompletionChunk, None, None]]:
         """
         Create a chat completion with ChatGPT API.
@@ -344,6 +435,8 @@ class Completions(BaseCompletions):
             stream: Whether to stream the response
             temperature: Sampling temperature (0-1)
             top_p: Nucleus sampling parameter (0-1)
+            tools: List of tool definitions available for the model to use
+            tool_choice: Control over which tool the model should use
             **kwargs: Additional parameters to pass to the API
 
         Returns:
@@ -362,9 +455,13 @@ class Completions(BaseCompletions):
                 max_tokens=max_tokens,
                 temperature=temperature,
                 top_p=top_p,
-                **kwargs
+                tools=tools,
+                tool_choice=tool_choice,
+                timeout=timeout,
+                proxies=proxies,
+                **kwargs,
             )
-        
+
         # Otherwise use non-streaming implementation
         return self._create_non_streaming(
             model=model,
@@ -372,7 +469,11 @@ class Completions(BaseCompletions):
             max_tokens=max_tokens,
             temperature=temperature,
             top_p=top_p,
-            **kwargs
+            tools=tools,
+            tool_choice=tool_choice,
+            timeout=timeout,
+            proxies=proxies,
+            **kwargs,
         )
 
     def _create_streaming(
@@ -383,62 +484,182 @@ class Completions(BaseCompletions):
         max_tokens: Optional[int] = None,
         temperature: Optional[float] = None,
         top_p: Optional[float] = None,
-        **kwargs: Any
+        tools: Optional[List[Union[Tool, Dict[str, Any]]]] = None,
+        tool_choice: Optional[Union[str, Dict[str, Any]]] = None,
+        timeout: Optional[int] = None,
+        proxies: Optional[dict] = None,
+        **kwargs: Any,
     ) -> Generator[ChatCompletionChunk, None, None]:
         """Implementation for streaming chat completions."""
         try:
             # Generate request ID and timestamp
             request_id = str(uuid.uuid4())
             created_time = int(time.time())
-            
+
             # Get the last user message
             last_user_message = None
             for msg in reversed(messages):
                 if msg["role"] == "user":
                     last_user_message = msg["content"]
                     break
-            
+
             if not last_user_message:
                 raise ValueError("No user message found in the conversation")
-            
-            # Get the response from ChatGPT
-            response = self._chatgpt_reversed.complete(last_user_message, model=model)
-            
-            # Split the response into chunks for streaming simulation
-            chunk_size = 10  # Characters per chunk
-            for i in range(0, len(response), chunk_size):
-                chunk_text = response[i:i+chunk_size]
-                
-                # Create and yield a chunk
-                delta = ChoiceDelta(content=chunk_text)
-                choice = Choice(index=0, delta=delta, finish_reason=None)
-                chunk = ChatCompletionChunk(
-                    id=request_id,
-                    choices=[choice],
-                    created=created_time,
-                    model=model
+
+            # Initialize ChatGPTReversed if not already initialized
+            if self._chatgpt_reversed is None:
+                self._chatgpt_reversed = ChatGPTReversed(model=model)
+
+            # Create a proper streaming request to ChatGPT
+            session_data = self._chatgpt_reversed.rotate_session_data()
+
+            headers = self._chatgpt_reversed.simulate_bypass_headers(
+                accept="text/event-stream", spoof_address=True, pre_oai_uuid=session_data["uuid"]
+            )
+
+            headers.update(
+                {
+                    "Cookie": f"__Host-next-auth.csrf-token={session_data['csrf']}; oai-did={session_data['uuid']}; oai-nav-state=1; oai-sc={session_data['sentinel']['oaiSc']};",
+                    "openai-sentinel-chat-requirements-token": session_data["sentinel"]["token"],
+                    "openai-sentinel-proof-token": session_data["sentinel"]["proof"],
+                }
+            )
+
+            # Format messages properly for ChatGPT
+            formatted_messages = []
+            for i, msg in enumerate(messages):
+                formatted_messages.append(
+                    {
+                        "id": str(uuid.uuid4()),
+                        "author": {"role": msg["role"]},
+                        "content": {"content_type": "text", "parts": [msg["content"]]},
+                        "metadata": {},
+                    }
                 )
-                
-                yield chunk
-                
-                # Add a small delay to simulate streaming
-                time.sleep(0.05)
-            
-            # Final chunk with finish_reason
-            delta = ChoiceDelta(content=None)
-            choice = Choice(index=0, delta=delta, finish_reason="stop")
-            chunk = ChatCompletionChunk(
-                id=request_id,
-                choices=[choice],
-                created=created_time,
-                model=model
+
+            payload = {
+                "action": "next",
+                "messages": formatted_messages,
+                "parent_message_id": str(uuid.uuid4()),
+                "model": model,
+                "timezone_offset_min": -120,
+                "suggestions": [],
+                "history_and_training_disabled": False,
+                "conversation_mode": {"kind": "primary_assistant", "plugin_ids": None},
+                "force_paragem": False,
+                "force_paragem_model_slug": "",
+                "force_nulligen": False,
+                "force_rate_limit": False,
+                "reset_rate_limits": False,
+                "websocket_request_id": str(uuid.uuid4()),
+                "force_use_sse": True,
+            }
+
+            # Add optional parameters if provided
+            if max_tokens is not None:
+                payload["max_tokens"] = max_tokens
+            if temperature is not None:
+                payload["temperature"] = temperature
+            if top_p is not None:
+                payload["top_p"] = top_p
+
+            # Make the actual streaming request
+            response = requests.post(
+                "https://chatgpt.com/backend-anon/conversation",
+                json=payload,
+                headers=headers,
+                stream=True,
+                timeout=timeout or 30,
             )
-            
-            yield chunk
-            
+
+            response.raise_for_status()
+
+            # Track conversation state
+            full_content = ""
+            prompt_tokens = count_tokens(str(messages))
+            completion_tokens = 0
+            total_tokens = prompt_tokens
+
+            # Process the streaming response
+            for line in response.iter_lines(decode_unicode=True):
+                if line:
+                    if line.startswith("data: "):
+                        data_str = line[6:]  # Remove "data: " prefix
+
+                        # Handle [DONE] message
+                        if data_str.strip() == "[DONE]":
+                            # Final chunk with finish_reason
+                            delta = ChoiceDelta(content=None)
+                            choice = Choice(index=0, delta=delta, finish_reason="stop")
+                            chunk = ChatCompletionChunk(
+                                id=request_id, choices=[choice], created=created_time, model=model
+                            )
+                            chunk.usage = {
+                                "prompt_tokens": prompt_tokens,
+                                "completion_tokens": completion_tokens,
+                                "total_tokens": total_tokens,
+                            }
+                            yield chunk
+                            break
+
+                        try:
+                            data = json.loads(data_str)
+
+                            # Handle different types of messages
+                            if data.get("message"):
+                                message = data["message"]
+
+                                # Handle assistant responses
+                                if message.get("author", {}).get("role") == "assistant":
+                                    content_parts = message.get("content", {}).get("parts", [])
+                                    if content_parts:
+                                        new_content = content_parts[0]
+
+                                        # Get the delta (new content since last chunk)
+                                        delta_content = (
+                                            new_content[len(full_content) :]
+                                            if new_content.startswith(full_content)
+                                            else new_content
+                                        )
+                                        full_content = new_content
+                                        completion_tokens = count_tokens(full_content)
+                                        total_tokens = prompt_tokens + completion_tokens
+
+                                        # Only yield chunk if there's new content
+                                        if delta_content:
+                                            delta = ChoiceDelta(
+                                                content=delta_content, role="assistant"
+                                            )
+                                            choice = Choice(
+                                                index=0, delta=delta, finish_reason=None
+                                            )
+                                            chunk = ChatCompletionChunk(
+                                                id=request_id,
+                                                choices=[choice],
+                                                created=created_time,
+                                                model=model,
+                                            )
+                                            chunk.usage = {
+                                                "prompt_tokens": prompt_tokens,
+                                                "completion_tokens": completion_tokens,
+                                                "total_tokens": total_tokens,
+                                            }
+                                            yield chunk
+
+                                # Handle finish status
+                                if message.get("status") == "finished_successfully":
+                                    pass
+
+                            elif data.get("type") == "message_stream_complete":
+                                # Stream is complete
+                                pass
+
+                        except json.JSONDecodeError:
+                            # Skip invalid JSON lines
+                            continue
+
         except Exception as e:
-            print(f"{RED}Error during ChatGPT streaming request: {e}{RESET}")
-            raise IOError(f"ChatGPT streaming request failed: {e}") from e
+            raise IOError(f"ChatGPT request failed: {e}") from e
 
     def _create_non_streaming(
         self,
@@ -448,50 +669,133 @@ class Completions(BaseCompletions):
         max_tokens: Optional[int] = None,
         temperature: Optional[float] = None,
         top_p: Optional[float] = None,
-        **kwargs: Any
+        tools: Optional[List[Union[Tool, Dict[str, Any]]]] = None,
+        tool_choice: Optional[Union[str, Dict[str, Any]]] = None,
+        timeout: Optional[int] = None,
+        proxies: Optional[dict] = None,
+        **kwargs: Any,
     ) -> ChatCompletion:
         """Implementation for non-streaming chat completions."""
         try:
             # Generate request ID and timestamp
             request_id = str(uuid.uuid4())
             created_time = int(time.time())
-            
-            # Get the last user message
-            last_user_message = None
-            for msg in reversed(messages):
-                if msg["role"] == "user":
-                    last_user_message = msg["content"]
-                    break
-            
-            if not last_user_message:
-                raise ValueError("No user message found in the conversation")
-            
-            # Get the response from ChatGPT
-            full_content = self._chatgpt_reversed.complete(last_user_message, model=model)
-            
-            # Create the completion message
-            message = ChatCompletionMessage(
-                role="assistant",
-                content=full_content
+
+            # Initialize ChatGPTReversed if not already initialized
+            if self._chatgpt_reversed is None:
+                self._chatgpt_reversed = ChatGPTReversed(model=model)
+
+            # Create a proper request to ChatGPT
+            session_data = self._chatgpt_reversed.rotate_session_data()
+
+            headers = self._chatgpt_reversed.simulate_bypass_headers(
+                accept="text/event-stream", spoof_address=True, pre_oai_uuid=session_data["uuid"]
             )
-            
-            # Create the choice
-            choice = Choice(
-                index=0,
-                message=message,
-                finish_reason="stop"
+
+            headers.update(
+                {
+                    "Cookie": f"__Host-next-auth.csrf-token={session_data['csrf']}; oai-did={session_data['uuid']}; oai-nav-state=1; oai-sc={session_data['sentinel']['oaiSc']};",
+                    "openai-sentinel-chat-requirements-token": session_data["sentinel"]["token"],
+                    "openai-sentinel-proof-token": session_data["sentinel"]["proof"],
+                }
             )
-            
-            # Estimate token usage (very rough estimate)
-            prompt_tokens = sum(len(msg.get("content", "")) // 4 for msg in messages)
-            completion_tokens = len(full_content) // 4
+
+            # Format messages properly for ChatGPT
+            formatted_messages = []
+            for i, msg in enumerate(messages):
+                formatted_messages.append(
+                    {
+                        "id": str(uuid.uuid4()),
+                        "author": {"role": msg["role"]},
+                        "content": {"content_type": "text", "parts": [msg["content"]]},
+                        "metadata": {},
+                    }
+                )
+
+            payload = {
+                "action": "next",
+                "messages": formatted_messages,
+                "parent_message_id": str(uuid.uuid4()),
+                "model": model,
+                "timezone_offset_min": -120,
+                "suggestions": [],
+                "history_and_training_disabled": False,
+                "conversation_mode": {"kind": "primary_assistant", "plugin_ids": None},
+                "force_paragem": False,
+                "force_paragem_model_slug": "",
+                "force_nulligen": False,
+                "force_rate_limit": False,
+                "reset_rate_limits": False,
+                "websocket_request_id": str(uuid.uuid4()),
+                "force_use_sse": True,
+            }
+
+            # Add optional parameters if provided
+            if max_tokens is not None:
+                payload["max_tokens"] = max_tokens
+            if temperature is not None:
+                payload["temperature"] = temperature
+            if top_p is not None:
+                payload["top_p"] = top_p
+
+            # Make the request and collect full response
+            response = requests.post(
+                "https://chatgpt.com/backend-anon/conversation",
+                json=payload,
+                headers=headers,
+                stream=True,
+                timeout=timeout or 30,
+            )
+
+            response.raise_for_status()
+
+            # Collect and parse the full response
+            full_response = ""
+            for line in response.iter_lines(decode_unicode=True):
+                if line:
+                    if line.startswith("data: "):
+                        data_str = line[6:]  # Remove "data: " prefix
+
+                        # Handle [DONE] message
+                        if data_str.strip() == "[DONE]":
+                            break
+
+                        try:
+                            data = json.loads(data_str)
+
+                            # Handle assistant responses
+                            if (
+                                data.get("message")
+                                and data["message"].get("author", {}).get("role") == "assistant"
+                            ):
+                                content_parts = data["message"].get("content", {}).get("parts", [])
+                                if content_parts:
+                                    full_response = content_parts[0]
+
+                        except json.JSONDecodeError:
+                            # Skip invalid JSON lines
+                            continue
+
+            # Create the completion message
+            message = ChatCompletionMessage(role="assistant", content=full_response)
+
+            # Create the choice
+            choice = Choice(index=0, message=message, finish_reason="stop")
+
+            # Calculate token usage using count_tokens
+            # Count tokens in the input messages (prompt)
+            prompt_tokens = count_tokens(str(messages))
+            # Count tokens in the response (completion)
+            completion_tokens = count_tokens(full_response)
+            total_tokens = prompt_tokens + completion_tokens
+
             usage = CompletionUsage(
                 prompt_tokens=prompt_tokens,
                 completion_tokens=completion_tokens,
-                total_tokens=prompt_tokens + completion_tokens
+                total_tokens=total_tokens,
             )
-            
-            # Create the completion object
+
+            # Create the completion object with correct OpenAI format
             completion = ChatCompletion(
                 id=request_id,
                 choices=[choice],
@@ -499,17 +803,19 @@ class Completions(BaseCompletions):
                 model=model,
                 usage=usage,
             )
-            
+
             return completion
-            
+
         except Exception as e:
             print(f"{RED}Error during ChatGPT non-stream request: {e}{RESET}")
             raise IOError(f"ChatGPT request failed: {e}") from e
 
+
 class Chat(BaseChat):
-    def __init__(self, client: 'ChatGPT'):
+    def __init__(self, client: "ChatGPT"):
         self.completions = Completions(client)
 
+
 class ChatGPT(OpenAICompatibleProvider):
     """
     OpenAI-compatible client for ChatGPT API.
@@ -523,27 +829,42 @@ class ChatGPT(OpenAICompatibleProvider):
         print(response.choices[0].message.content)
     """
 
-    AVAILABLE_MODELS = [
-        "auto", 
-        "gpt-4o-mini", 
-        "gpt-4o", 
-        "o4-mini"
-    ]
+    required_auth = False
 
     def __init__(
         self,
-        timeout: int = 60,
-        proxies: dict = {}
+        api_key: Optional[str] = None,
+        tools: Optional[List[Tool]] = None,
+        proxies: Optional[Dict[str, str]] = None,
     ):
         """
         Initialize the ChatGPT client.
 
         Args:
-            timeout: Request timeout in seconds
-            proxies: Optional proxy configuration
+            api_key: Optional API key (not used by ChatGPTReversed but included for interface compatibility)
+            tools: Optional list of tools to register with the provider
+            proxies: Optional proxy configuration dict, e.g. {"http": "http://proxy:8080", "https": "https://proxy:8080"}
         """
-        self.timeout = timeout
-        self.proxies = proxies
-        
+        super().__init__(api_key=api_key, tools=tools, proxies=proxies)
         # Initialize chat interface
         self.chat = Chat(self)
+
+    @property
+    def AVAILABLE_MODELS(self):
+        return ChatGPTReversed.AVAILABLE_MODELS
+
+    @property
+    def models(self) -> SimpleModelList:
+        return SimpleModelList(self.AVAILABLE_MODELS)
+
+
+if __name__ == "__main__":
+    # Example usage
+    client = ChatGPT()
+    response = client.chat.completions.create(
+        model="o4-mini-high", messages=[{"role": "user", "content": "How many r in strawberry"}]
+    )
+    if isinstance(response, ChatCompletion):
+        if response.choices[0].message and response.choices[0].message.content:
+            print(response.choices[0].message.content)
+    print()