Merge branch 'develop' of https://github.com/unclecode/crawl4ai into develop

Author: ntohidi

deploy/docker/README.md

@@ -692,8 +692,7 @@ app:
 # Default LLM Configuration
 llm:
   provider: "openai/gpt-4o-mini"  # Can be overridden by LLM_PROVIDER env var
-  api_key_env: "OPENAI_API_KEY"
-  # api_key: sk-...  # If you pass the API key directly then api_key_env will be ignored
+  # api_key: sk-...  # If you pass the API key directly (not recommended)
 
 # Redis Configuration (Used by internal Redis server managed by supervisord)
 redis:

deploy/docker/api.py

@@ -96,7 +96,7 @@ async def handle_llm_qa(
         response = perform_completion_with_backoff(
             provider=config["llm"]["provider"],
             prompt_with_variables=prompt,
-            api_token=get_llm_api_key(config)
+            api_token=get_llm_api_key(config)  # Returns None to let litellm handle it
         )
 
         return response.choices[0].message.content
@@ -127,7 +127,7 @@ async def process_llm_extraction(
                 "error": error_msg
             })
             return
-        api_key = get_llm_api_key(config, provider)
+        api_key = get_llm_api_key(config, provider)  # Returns None to let litellm handle it
         llm_strategy = LLMExtractionStrategy(
             llm_config=LLMConfig(
                 provider=provider or config["llm"]["provider"],
@@ -203,7 +203,7 @@ async def handle_markdown_request(
                 FilterType.LLM: LLMContentFilter(
                     llm_config=LLMConfig(
                         provider=provider or config["llm"]["provider"],
-                        api_token=get_llm_api_key(config, provider),
+                        api_token=get_llm_api_key(config, provider),  # Returns None to let litellm handle it
                     ),
                     instruction=query or "Extract main content"
                 )

deploy/docker/config.yml

@@ -11,8 +11,7 @@ app:
 # Default LLM Configuration
 llm:
   provider: "openai/gpt-4o-mini"
-  api_key_env: "OPENAI_API_KEY"
-  # api_key: sk-...  # If you pass the API key directly then api_key_env will be ignored
+  # api_key: sk-...  # If you pass the API key directly (not recommended)
 
 # Redis Configuration
 redis:

deploy/docker/utils.py

@@ -71,27 +71,22 @@ def decode_redis_hash(hash_data: Dict[bytes, bytes]) -> Dict[str, str]:
 
 
 
-def get_llm_api_key(config: Dict, provider: Optional[str] = None) -> str:
+def get_llm_api_key(config: Dict, provider: Optional[str] = None) -> Optional[str]:
     """Get the appropriate API key based on the LLM provider.
     
     Args:
         config: The application configuration dictionary
         provider: Optional provider override (e.g., "openai/gpt-4")
     
     Returns:
-        The API key for the provider, or empty string if not found
+        The API key if directly configured, otherwise None to let litellm handle it
     """
-        
-    # Use provided provider or fall back to config
-    if not provider:
-        provider = config["llm"]["provider"]
-    
-    # Check if direct API key is configured
+    # Check if direct API key is configured (for backward compatibility)
     if "api_key" in config["llm"]:
         return config["llm"]["api_key"]
 
-    # Fall back to the configured api_key_env if no match
-    return os.environ.get(config["llm"].get("api_key_env", ""), "")
+    # Return None - litellm will automatically find the right environment variable
+    return None
 
 
 def validate_llm_provider(config: Dict, provider: Optional[str] = None) -> tuple[bool, str]:
@@ -104,16 +99,12 @@ def validate_llm_provider(config: Dict, provider: Optional[str] = None) -> tuple
     Returns:
         Tuple of (is_valid, error_message)
     """
-    # Use provided provider or fall back to config
-    if not provider:
-        provider = config["llm"]["provider"]
-    
-    # Get the API key for this provider
-    api_key = get_llm_api_key(config, provider)
-    
-    if not api_key:
-        return False, f"No API key found for provider '{provider}'. Please set the appropriate environment variable."
+    # If a direct API key is configured, validation passes
+    if "api_key" in config["llm"]:
+        return True, ""
 
+    # Otherwise, trust that litellm will find the appropriate environment variable
+    # We can't easily validate this without reimplementing litellm's logic
     return True, ""
 
 

docs/md_v2/core/docker-deployment.md

@@ -176,7 +176,7 @@ The Docker setup now supports flexible LLM provider configuration through three
 
 3. **Config File Default**: Falls back to `config.yml` (default: `openai/gpt-4o-mini`)
 
-The system automatically selects the appropriate API key based on the configured `api_key_env` in the config file.
+The system automatically selects the appropriate API key based on the provider. LiteLLM handles finding the correct environment variable for each provider (e.g., OPENAI_API_KEY for OpenAI, GEMINI_API_TOKEN for Google Gemini, etc.).
 
 #### 3. Build and Run with Compose
 
@@ -693,8 +693,7 @@ app:
 # Default LLM Configuration
 llm:
   provider: "openai/gpt-4o-mini"  # Can be overridden by LLM_PROVIDER env var
-  api_key_env: "OPENAI_API_KEY"
-  # api_key: sk-...  # If you pass the API key directly then api_key_env will be ignored
+  # api_key: sk-...  # If you pass the API key directly (not recommended)
 
 # Redis Configuration (Used by internal Redis server managed by supervisord)
 redis: