just why not

tools/ai-review/clients/llm_client.py

@@ -77,11 +77,13 @@ class OpenAIProvider(BaseLLMProvider):
         model: str = "gpt-4o-mini",
         temperature: float = 0,
         max_tokens: int = 4096,
+        timeout: int = 120,
     ):
         self.api_key = api_key or os.environ.get("OPENAI_API_KEY", "")
         self.model = model
         self.temperature = temperature
         self.max_tokens = max_tokens
+        self.timeout = timeout
         self.api_url = "https://api.openai.com/v1/chat/completions"
 
     def call(self, prompt: str, **kwargs) -> LLMResponse:
@@ -101,7 +103,7 @@ class OpenAIProvider(BaseLLMProvider):
                 "max_tokens": kwargs.get("max_tokens", self.max_tokens),
                 "messages": [{"role": "user", "content": prompt}],
             },
-            timeout=120,
+            timeout=self.timeout,
         )
         response.raise_for_status()
         data = response.json()
@@ -145,7 +147,7 @@ class OpenAIProvider(BaseLLMProvider):
                 "Content-Type": "application/json",
             },
             json=request_body,
-            timeout=120,
+            timeout=self.timeout,
         )
         response.raise_for_status()
         data = response.json()
@@ -186,11 +188,13 @@ class OpenRouterProvider(BaseLLMProvider):
         model: str = "anthropic/claude-3.5-sonnet",
         temperature: float = 0,
         max_tokens: int = 4096,
+        timeout: int = 120,
     ):
         self.api_key = api_key or os.environ.get("OPENROUTER_API_KEY", "")
         self.model = model
         self.temperature = temperature
         self.max_tokens = max_tokens
+        self.timeout = timeout
         self.api_url = "https://openrouter.ai/api/v1/chat/completions"
 
     def call(self, prompt: str, **kwargs) -> LLMResponse:
@@ -210,7 +214,7 @@ class OpenRouterProvider(BaseLLMProvider):
                 "max_tokens": kwargs.get("max_tokens", self.max_tokens),
                 "messages": [{"role": "user", "content": prompt}],
             },
-            timeout=120,
+            timeout=self.timeout,
         )
         response.raise_for_status()
         data = response.json()
@@ -254,7 +258,7 @@ class OpenRouterProvider(BaseLLMProvider):
                 "Content-Type": "application/json",
             },
             json=request_body,
-            timeout=120,
+            timeout=self.timeout,
         )
         response.raise_for_status()
         data = response.json()
@@ -294,10 +298,12 @@ class OllamaProvider(BaseLLMProvider):
         host: str | None = None,
         model: str = "codellama:13b",
         temperature: float = 0,
+        timeout: int = 300,
     ):
         self.host = host or os.environ.get("OLLAMA_HOST", "http://localhost:11434")
         self.model = model
         self.temperature = temperature
+        self.timeout = timeout
 
     def call(self, prompt: str, **kwargs) -> LLMResponse:
         """Call Ollama API."""
@@ -311,7 +317,7 @@ class OllamaProvider(BaseLLMProvider):
                     "temperature": kwargs.get("temperature", self.temperature),
                 },
             },
-            timeout=300,  # Longer timeout for local models
+            timeout=self.timeout,
         )
         response.raise_for_status()
         data = response.json()
@@ -477,12 +483,18 @@ class LLMClient:
         provider = config.get("provider", "openai")
         provider_config = {}
 
+        # Get timeout configuration
+        timeouts = config.get("timeouts", {})
+        llm_timeout = timeouts.get("llm", 120)
+        ollama_timeout = timeouts.get("ollama", 300)
+
         # Map config keys to provider-specific settings
         if provider == "openai":
             provider_config = {
                 "model": config.get("model", {}).get("openai", "gpt-4o-mini"),
                 "temperature": config.get("temperature", 0),
                 "max_tokens": config.get("max_tokens", 16000),
+                "timeout": llm_timeout,
             }
         elif provider == "openrouter":
             provider_config = {
@@ -491,11 +503,13 @@ class LLMClient:
                 ),
                 "temperature": config.get("temperature", 0),
                 "max_tokens": config.get("max_tokens", 16000),
+                "timeout": llm_timeout,
             }
         elif provider == "ollama":
             provider_config = {
                 "model": config.get("model", {}).get("ollama", "codellama:13b"),
                 "temperature": config.get("temperature", 0),
+                "timeout": ollama_timeout,
             }
 
         return cls(provider=provider, config=provider_config)