webscout 8.0__py3-none-any.whl → 8.1__py3-none-any.whl

webscout/Provider/OPENAI/exachat.py

@@ -0,0 +1,421 @@
+import time
+import uuid
+import requests
+import json
+from typing import List, Dict, Optional, Union, Generator, Any
+
+from webscout.litagent import LitAgent
+from .base import BaseChat, BaseCompletions, OpenAICompatibleProvider
+from .utils import (
+    ChatCompletion,
+    ChatCompletionChunk,
+    Choice,
+    ChatCompletionMessage,
+    ChoiceDelta,
+    CompletionUsage,
+    format_prompt
+)
+
+# ANSI escape codes for formatting
+BOLD = "\033[1m"
+RED = "\033[91m"
+RESET = "\033[0m"
+
+# Model configurations
+MODEL_CONFIGS = {
+    "exaanswer": {
+        "endpoint": "https://exa-chat.vercel.app/api/exaanswer",
+        "models": ["exaanswer"],
+    },
+    "gemini": {
+        "endpoint": "https://exa-chat.vercel.app/api/gemini",
+        "models": [
+            "gemini-2.0-flash",
+            "gemini-2.0-flash-exp-image-generation",
+            "gemini-2.0-flash-thinking-exp-01-21",
+            "gemini-2.5-pro-exp-03-25",
+            "gemini-2.0-pro-exp-02-05",
+        ],
+    },
+    "openrouter": {
+        "endpoint": "https://exa-chat.vercel.app/api/openrouter",
+        "models": [
+            "mistralai/mistral-small-3.1-24b-instruct:free",
+            "deepseek/deepseek-r1:free",
+            "deepseek/deepseek-chat-v3-0324:free",
+            "google/gemma-3-27b-it:free",
+            "meta-llama/llama-4-maverick:free",           
+        ],
+    },
+    "groq": {
+        "endpoint": "https://exa-chat.vercel.app/api/groq",
+        "models": [
+            "deepseek-r1-distill-llama-70b",
+            "deepseek-r1-distill-qwen-32b",
+            "gemma2-9b-it",
+            "llama-3.1-8b-instant",
+            "llama-3.2-1b-preview",
+            "llama-3.2-3b-preview",
+            "llama-3.2-90b-vision-preview",
+            "llama-3.3-70b-specdec",
+            "llama-3.3-70b-versatile",
+            "llama3-70b-8192",
+            "llama3-8b-8192",
+            "qwen-2.5-32b",
+            "qwen-2.5-coder-32b",
+            "qwen-qwq-32b",
+            "meta-llama/llama-4-scout-17b-16e-instruct"
+        ],
+    },
+    "cerebras": {
+        "endpoint": "https://exa-chat.vercel.app/api/cerebras",
+        "models": [
+            "llama3.1-8b",
+            "llama-3.3-70b"
+        ],
+    },
+}
+
+class Completions(BaseCompletions):
+    def __init__(self, client: 'ExaChat'):
+        self._client = client
+
+    def create(
+        self,
+        *,
+        model: str,
+        messages: List[Dict[str, str]],
+        max_tokens: Optional[int] = None,  # Not used directly but kept for compatibility
+        stream: bool = False,
+        temperature: Optional[float] = None,
+        top_p: Optional[float] = None,
+        **kwargs: Any
+    ) -> Union[ChatCompletion, Generator[ChatCompletionChunk, None, None]]:
+        """
+        Creates a model response for the given chat conversation.
+        Mimics openai.chat.completions.create
+        """
+        # Format the messages using the format_prompt utility
+        # This creates a conversation in the format: "User: message\nAssistant: response\nUser: message\nAssistant:"
+        question = format_prompt(messages, add_special_tokens=True, do_continue=True)
+        
+        # Determine the provider based on the model
+        provider = self._client._get_provider_from_model(model)
+        
+        # Build the appropriate payload based on the provider
+        if provider == "exaanswer":
+            payload = {
+                "query": question,
+                "messages": []
+            }
+        elif provider in ["gemini", "cerebras"]:
+            payload = {
+                "query": question,
+                "model": model,
+                "messages": []
+            }
+        else:  # openrouter or groq
+            payload = {
+                "query": question + "\n",  # Add newline for openrouter and groq models
+                "model": model,
+                "messages": []
+            }
+
+        request_id = f"chatcmpl-{uuid.uuid4()}"
+        created_time = int(time.time())
+
+        if stream:
+            return self._create_stream(request_id, created_time, model, provider, payload)
+        else:
+            return self._create_non_stream(request_id, created_time, model, provider, payload)
+
+    def _create_stream(
+        self, request_id: str, created_time: int, model: str, provider: str, payload: Dict[str, Any]
+    ) -> Generator[ChatCompletionChunk, None, None]:
+        try:
+            endpoint = self._client._get_endpoint(provider)
+            response = self._client.session.post(
+                endpoint,
+                headers=self._client.headers,
+                json=payload,
+                stream=True,
+                timeout=self._client.timeout
+            )
+            response.raise_for_status()
+
+            # Track token usage across chunks
+            completion_tokens = 0
+            streaming_text = ""
+
+            for line in response.iter_lines():
+                if not line:
+                    continue
+                
+                try:
+                    data = json.loads(line.decode('utf-8'))
+                    if 'choices' in data and len(data['choices']) > 0:
+                        content = data['choices'][0].get('delta', {}).get('content', '')
+                        if content:
+                            streaming_text += content
+                            completion_tokens += len(content) // 4  # Rough estimate
+                            
+                            # Create a delta object for this chunk
+                            delta = ChoiceDelta(content=content)
+                            choice = Choice(index=0, delta=delta, finish_reason=None)
+                            
+                            chunk = ChatCompletionChunk(
+                                id=request_id,
+                                choices=[choice],
+                                created=created_time,
+                                model=model,
+                            )
+                            
+                            yield chunk
+                except json.JSONDecodeError:
+                    continue
+
+            # 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,
+            )
+            
+            yield chunk
+
+        except requests.exceptions.RequestException as e:
+            print(f"{RED}Error during ExaChat stream request: {e}{RESET}")
+            raise IOError(f"ExaChat request failed: {e}") from e
+
+    def _create_non_stream(
+        self, request_id: str, created_time: int, model: str, provider: str, payload: Dict[str, Any]
+    ) -> ChatCompletion:
+        try:
+            endpoint = self._client._get_endpoint(provider)
+            response = self._client.session.post(
+                endpoint,
+                headers=self._client.headers,
+                json=payload,
+                timeout=self._client.timeout
+            )
+            response.raise_for_status()
+
+            full_response = ""
+            for line in response.iter_lines():
+                if line:
+                    try:
+                        data = json.loads(line.decode('utf-8'))
+                        if 'choices' in data and len(data['choices']) > 0:
+                            content = data['choices'][0].get('delta', {}).get('content', '')
+                            if content:
+                                full_response += content
+                    except json.JSONDecodeError:
+                        continue
+
+            # Create usage statistics (estimated)
+            prompt_tokens = len(payload["query"]) // 4
+            completion_tokens = len(full_response) // 4
+            total_tokens = prompt_tokens + completion_tokens
+            
+            usage = CompletionUsage(
+                prompt_tokens=prompt_tokens,
+                completion_tokens=completion_tokens,
+                total_tokens=total_tokens
+            )
+            
+            # Create the message object
+            message = ChatCompletionMessage(
+                role="assistant",
+                content=full_response
+            )
+            
+            # Create the choice object
+            choice = Choice(
+                index=0,
+                message=message,
+                finish_reason="stop"
+            )
+            
+            # Create the completion object
+            completion = ChatCompletion(
+                id=request_id,
+                choices=[choice],
+                created=created_time,
+                model=model,
+                usage=usage,
+            )
+            
+            return completion
+
+        except Exception as e:
+            print(f"{RED}Error during ExaChat non-stream request: {e}{RESET}")
+            raise IOError(f"ExaChat request failed: {e}") from e
+
+class Chat(BaseChat):
+    def __init__(self, client: 'ExaChat'):
+        self.completions = Completions(client)
+
+class ExaChat(OpenAICompatibleProvider):
+    """
+    OpenAI-compatible client for ExaChat API.
+
+    Usage:
+        client = ExaChat()
+        response = client.chat.completions.create(
+            model="exaanswer",
+            messages=[{"role": "user", "content": "Hello!"}]
+        )
+        print(response.choices[0].message.content)
+    """
+
+    AVAILABLE_MODELS = [
+        # ExaAnswer Models
+        "exaanswer",
+        
+        # Gemini Models
+        "gemini-2.0-flash",
+        "gemini-2.0-flash-exp-image-generation",
+        "gemini-2.0-flash-thinking-exp-01-21",
+        "gemini-2.5-pro-exp-03-25",
+        "gemini-2.0-pro-exp-02-05",
+        
+        # OpenRouter Models
+        "mistralai/mistral-small-3.1-24b-instruct:free",
+        "deepseek/deepseek-r1:free",
+        "deepseek/deepseek-chat-v3-0324:free",
+        "google/gemma-3-27b-it:free",
+        "meta-llama/llama-4-maverick:free",
+        
+        # Groq Models
+        "deepseek-r1-distill-llama-70b",
+        "deepseek-r1-distill-qwen-32b",
+        "gemma2-9b-it",
+        "llama-3.1-8b-instant",
+        "llama-3.2-1b-preview",
+        "llama-3.2-3b-preview",
+        "llama-3.2-90b-vision-preview",
+        "llama-3.3-70b-specdec",
+        "llama-3.3-70b-versatile",
+        "llama3-70b-8192",
+        "llama3-8b-8192",
+        "qwen-2.5-32b",
+        "qwen-2.5-coder-32b",
+        "qwen-qwq-32b",
+        "meta-llama/llama-4-scout-17b-16e-instruct",
+        
+        # Cerebras Models
+        "llama3.1-8b",
+        "llama-3.3-70b"
+    ]
+
+    def __init__(
+        self,
+        timeout: int = 30,
+        temperature: float = 0.5,
+        top_p: float = 1.0
+    ):
+        """
+        Initialize the ExaChat client.
+
+        Args:
+            timeout: Request timeout in seconds.
+            temperature: Temperature for response generation.
+            top_p: Top-p sampling parameter.
+        """
+        self.timeout = timeout
+        self.temperature = temperature
+        self.top_p = top_p
+        
+        # Initialize LitAgent for user agent generation
+        agent = LitAgent()
+        
+        self.headers = {
+            "accept": "*/*",
+            "accept-language": "en-US,en;q=0.9",
+            "content-type": "application/json",
+            "origin": "https://exa-chat.vercel.app",
+            "referer": "https://exa-chat.vercel.app/",
+            "user-agent": agent.random(),
+        }
+        
+        self.session = requests.Session()
+        self.session.headers.update(self.headers)
+        self.session.cookies.update({"session": uuid.uuid4().hex})
+        
+        # Initialize the chat interface
+        self.chat = Chat(self)
+
+    def _get_endpoint(self, provider: str) -> str:
+        """Get the API endpoint for the specified provider."""
+        return MODEL_CONFIGS[provider]["endpoint"]
+
+    def _get_provider_from_model(self, model: str) -> str:
+        """Determine the provider based on the model name."""
+        for provider, config in MODEL_CONFIGS.items():
+            if model in config["models"]:
+                return provider
+        
+        # If model not found, use a default model
+        print(f"{BOLD}Warning: Model '{model}' not found, using default model 'exaanswer'{RESET}")
+        return "exaanswer"
+
+    def convert_model_name(self, model: str) -> str:
+        """
+        Ensure the model name is in the correct format.
+        """
+        if model in self.AVAILABLE_MODELS:
+            return model
+        
+        # Try to find a matching model
+        for available_model in self.AVAILABLE_MODELS:
+            if model.lower() in available_model.lower():
+                return available_model
+        
+        # Default to exaanswer if no match
+        print(f"{BOLD}Warning: Model '{model}' not found, using default model 'exaanswer'{RESET}")
+        return "exaanswer"
+
+
+# Simple test if run directly
+if __name__ == "__main__":
+    print("-" * 80)
+    print(f"{'Model':<50} {'Status':<10} {'Response'}")
+    print("-" * 80)
+
+    # Test a subset of models to avoid excessive API calls
+    test_models = [
+        "exaanswer",
+        "gemini-2.0-flash",
+        "deepseek/deepseek-r1:free",
+        "llama-3.1-8b-instant",
+        "llama3.1-8b"
+    ]
+
+    for model in test_models:
+        try:
+            client = ExaChat(timeout=60)
+            # Test with a simple conversation to demonstrate format_prompt usage
+            response = client.chat.completions.create(
+                model=model,
+                messages=[
+                    {"role": "system", "content": "You are a helpful assistant."},
+                    {"role": "user", "content": "Say 'Hello' in one word"},
+                ],
+                stream=False
+            )
+            
+            if response and response.choices and response.choices[0].message.content:
+                status = "✓"
+                # Truncate response if too long
+                display_text = response.choices[0].message.content.strip()
+                display_text = display_text[:50] + "..." if len(display_text) > 50 else display_text
+            else:
+                status = "✗"
+                display_text = "Empty or invalid response"
+            print(f"{model:<50} {status:<10} {display_text}")
+        except Exception as e:
+            print(f"{model:<50} {'✗':<10} {str(e)}")