feat: Add Anthropic (Claude) LLM support

agent_assembly_line/llm_factory.py

@@ -26,6 +26,20 @@
             "embeddings": "text-embedding-ada-002"
         }
     },
+    "anthropic": {
+        "claude-3-5-sonnet": {
+            "llm": "claude-3-5-sonnet-20241022",
+            "embeddings": "text-embedding-ada-002"  # Use OpenAI for embeddings
+        },
+        "claude-3-5-haiku": {
+            "llm": "claude-3-5-haiku-20241022",
+            "embeddings": "text-embedding-ada-002"
+        },
+        "claude-3-opus": {
+            "llm": "claude-3-opus-20240229",
+            "embeddings": "text-embedding-ada-002"
+        }
+    },
     # Add more mappings as needed
 }
 
@@ -60,6 +74,54 @@ def create_llm_and_embeddings(config: Config):
             embeddings = OpenAIEmbeddings(api_key=api_key, model=embeddings)
             return llm, embeddings
 
+        elif llm_type == "anthropic":
+            from langchain_anthropic import ChatAnthropic
+            from langchain_openai.embeddings import OpenAIEmbeddings
+
+            api_key = os.getenv("ANTHROPIC_API_KEY")
+            if not api_key:
+                raise ValueError("ANTHROPIC_API_KEY not found in environment variables.")
+
+            # Map short names to full model IDs
+            model_mapping = {
+                "claude-3-5-sonnet": "claude-3-5-sonnet-20241022",
+                "claude-3-5-haiku": "claude-3-5-haiku-20241022",
+                "claude-3-opus": "claude-3-opus-20240229",
+                "claude-3-sonnet": "claude-3-sonnet-20240229",
+                "claude-3-haiku": "claude-3-haiku-20240307",
+            }
+
+            full_model_name = model_mapping.get(model_name, model_name)
+
+            llm = ChatAnthropic(
+                api_key=api_key,
+                model=full_model_name,
+                timeout=config.timeout,
+                max_tokens=config.max_tokens if hasattr(config, 'max_tokens') else 4096
+            )
+
+            # Anthropic doesn't provide embeddings, so use OpenAI or custom
+            openai_key = os.getenv("OPENAI_API_KEY")
+            if config.custom_embeddings:
+                embeddings_model = config.custom_embeddings
+            else:
+                embeddings_model = _llm_embeddings_mapping.get("anthropic", {}).get(model_name, {}).get("embeddings", "text-embedding-ada-002")
+
+            if openai_key:
+                embeddings = OpenAIEmbeddings(api_key=openai_key, model=embeddings_model)
+            else:
+                # Fallback to Ollama embeddings if available
+                try:
+                    from langchain_ollama.embeddings import OllamaEmbeddings
+                    embeddings = OllamaEmbeddings(model="nomic-embed-text")
+                except ImportError:
+                    raise ValueError(
+                        "Anthropic does not provide embeddings. "
+                        "Please set OPENAI_API_KEY for OpenAI embeddings or install langchain_ollama."
+                    )
+
+            return llm, embeddings
+
         elif llm_type == "runpod":
             from langchain_runpod.llms import RunpodLLM
             from langchain_runpod.embeddings import RunpodEmbeddings
@@ -87,4 +149,7 @@ def extract_response(response, config: Config):
                 return response["choices"][0]["text"]
         if config.model_name == "gpt-4o":
             return response["choices"][0]["text"]
+        # Handle Anthropic response format
+        if hasattr(response, 'content'):
+            return response.content
         return response

pyproject.toml

@@ -21,6 +21,7 @@ dependencies = [
     "langchain-chroma==0.2.2",
     "langchain-ollama==0.3.0",
     "langchain-openai==0.3.11",
+    "langchain-anthropic==0.3.10",
     "pandas==2.2.3",
     "numpy==1.26.4",
     "pytest==8.3.5",