langgraph-agent-toolkit 0.8.0__tar.gz → 0.8.2__tar.gz

{langgraph_agent_toolkit-0.8.0 → langgraph_agent_toolkit-0.8.2}/CHANGELOG.md

@@ -6,6 +6,22 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to
 [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.8.2]
+
+### Added
+
+- Factory for embedding models
+
+## [0.8.1]
+
+### Fixed
+
+- langfuse `score` -> `create_score`
+
+### Updated
+
+- Core Dependencies
+
 ## [0.8.0]
 
 ### Added

{langgraph_agent_toolkit-0.8.0 → langgraph_agent_toolkit-0.8.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: langgraph-agent-toolkit
-Version: 0.8.0
+Version: 0.8.2
 Summary: Full toolkit for running an AI agent service built with LangGraph, FastAPI and Streamlit
 Project-URL: documentation, https://kryvokhyzha.github.io/langgraph-agent-toolkit
 Project-URL: repository, https://github.com/kryvokhyzha/langgraph-agent-toolkit
@@ -23,36 +23,36 @@ Classifier: Topic :: Software Development :: Libraries :: Python Modules
 Requires-Python: <3.14,>=3.11
 Requires-Dist: ddgs>=9.5.0
 Requires-Dist: duckduckgo-search>=7.3.0
-Requires-Dist: fastapi<0.120.0,>=0.116.1
+Requires-Dist: fastapi<0.122.0,>=0.116.1
 Requires-Dist: fire~=0.7.0
 Requires-Dist: grpcio>=1.68.0
 Requires-Dist: httpx~=0.28.1
 Requires-Dist: jinja2~=3.1.6
-Requires-Dist: jiter<0.12,>=0.10
+Requires-Dist: jiter<0.13,>=0.10
 Requires-Dist: joblib>=1.4.2
 Requires-Dist: langchain-community<0.4.2,>=0.4.0
-Requires-Dist: langchain-core<1.0.3,>=1.0.0
-Requires-Dist: langchain<1.0.4,>=1.0.0
-Requires-Dist: langgraph-checkpoint-postgres<3.0.1,>=3.0.0
+Requires-Dist: langchain-core<1.0.5,>=1.0.0
+Requires-Dist: langchain<1.0.6,>=1.0.0
+Requires-Dist: langgraph-checkpoint-postgres<3.0.2,>=3.0.0
 Requires-Dist: langgraph-checkpoint-sqlite<3.0.1,>=3.0.0
 Requires-Dist: langgraph-cli[inmem]<0.5.0,>=0.2.10
 Requires-Dist: langgraph-supervisor~=0.0.21
-Requires-Dist: langgraph<1.0.3,>=1.0.0
+Requires-Dist: langgraph<1.0.4,>=1.0.0
 Requires-Dist: loguru~=0.7.3
 Requires-Dist: multidict>6.3.2
+Requires-Dist: numpy<2.4.0,>=1.26.4; python_version == '3.11'
 Requires-Dist: numpy<2.4.0,>=2.0.0; python_version >= '3.12'
-Requires-Dist: numpy~=1.26.4; python_version == '3.11'
-Requires-Dist: onnxruntime<1.23.0,>=1.20.0
+Requires-Dist: onnxruntime<1.24.0,>=1.20.0
 Requires-Dist: pandas<2.4.0,>=2.2.3
 Requires-Dist: psycopg[binary,pool]~=3.2.4
 Requires-Dist: pyarrow>=19.0.1
-Requires-Dist: pydantic-settings<2.12.0,>=2.9.1
+Requires-Dist: pydantic-settings<2.13.0,>=2.9.1
 Requires-Dist: pydantic<2.13.0,>=2.11.1
 Requires-Dist: pyowm<3.6,>=3.3
-Requires-Dist: python-dotenv~=1.1.0
+Requires-Dist: python-dotenv<1.3,>=1.1
 Requires-Dist: rootutils>=1.0.7
 Requires-Dist: setuptools>=75.6.0
-Requires-Dist: streamlit<1.51,>=1.45
+Requires-Dist: streamlit<1.52,>=1.45
 Requires-Dist: uvicorn<0.39,>=0.35
 Requires-Dist: watchdog~=6.0.0
 Provides-Extra: all-backends
@@ -70,7 +70,7 @@ Requires-Dist: langchain-groq~=1.0.0; extra == 'all-llms'
 Requires-Dist: langchain-ollama~=1.0.0; extra == 'all-llms'
 Requires-Dist: langchain-openai~=1.0.0; extra == 'all-llms'
 Provides-Extra: all-observability
-Requires-Dist: langfuse<3.8.2,>=2.60.3; extra == 'all-observability'
+Requires-Dist: langfuse<3.9.2,>=2.60.3; extra == 'all-observability'
 Requires-Dist: langsmith<0.5.0,>=0.3.31; extra == 'all-observability'
 Provides-Extra: anthropic
 Requires-Dist: langchain-anthropic~=1.0.0; extra == 'anthropic'
@@ -92,13 +92,13 @@ Provides-Extra: gunicorn-backend
 Requires-Dist: gunicorn~=23.0.0; extra == 'gunicorn-backend'
 Requires-Dist: uvicorn[standard]<0.39,>=0.35; extra == 'gunicorn-backend'
 Provides-Extra: langfuse
-Requires-Dist: langfuse<3.8.2,>=2.60.3; extra == 'langfuse'
+Requires-Dist: langfuse<3.9.2,>=2.60.3; extra == 'langfuse'
 Provides-Extra: langsmith
 Requires-Dist: langsmith<0.5.0,>=0.3.31; extra == 'langsmith'
 Provides-Extra: openai
 Requires-Dist: langchain-openai~=1.0.0; extra == 'openai'
 Provides-Extra: uvicorn-backend
-Requires-Dist: uvicorn~=0.35.0; extra == 'uvicorn-backend'
+Requires-Dist: uvicorn<0.39,>=0.35; extra == 'uvicorn-backend'
 Description-Content-Type: text/markdown
 
 <div align="center">

{langgraph_agent_toolkit-0.8.0 → langgraph_agent_toolkit-0.8.2}/docs/conf.py

@@ -112,7 +112,7 @@ autosummary_generate = True
 # Intersphinx mappings - update with corrected URLs
 intersphinx_mapping = {
     "python": ("https://docs.python.org/3", None),
-    "langgraph": ("https://python.langchain.com/docs/integrations", None),  # Fixed URL
+    "langgraph": ("https://langchain-ai.github.io/langgraph/", None),
     "pydantic": ("https://docs.pydantic.dev/latest/", None),
     "fastapi": ("https://fastapi.tiangolo.com/", None),
 }

{langgraph_agent_toolkit-0.8.0 → langgraph_agent_toolkit-0.8.2}/langgraph_agent_toolkit/agents/agent_executor.py

@@ -316,36 +316,43 @@ class AgentExecutor:
             recursion_limit=recursion_limit,
         )
 
-        # Invoke the agent
-        response_events: list[tuple[str, Any]] = await agent.graph.ainvoke(
+        # Wrap execution in trace context
+        with agent.observability.trace_context(
+            run_id=run_id,
+            user_id=user_id,
             input=input_data,
-            config=config,
-            # stream_mode=["updates", "values"],
-            stream_mode=["values"],
-        )
+            agent_name=agent.name,
+        ):
+            # Invoke the agent
+            response_events: list[tuple[str, Any]] = await agent.graph.ainvoke(
+                input=input_data,
+                config=config,
+                # stream_mode=["updates", "values"],
+                stream_mode=["values"],
+            )
 
-        response_type, response = response_events[-1]
-
-        if response_type == "values" and "__interrupt__" not in response:
-            generated_message = response.get("structured_response")
-            if not generated_message:
-                generated_message = response["messages"][-1]
-
-            # Normal response, the agent completed successfully
-            output = langchain_to_chat_message(generated_message)
-        elif response_type == "values" and "__interrupt__" in response:
-            # The last thing to occur was an interrupt
-            # Return the value of the first interrupt as an AIMessage
-            output = langchain_to_chat_message(AIMessage(content=response["__interrupt__"][0].value))
-        elif response_type == "updates" and "__interrupt__" in response:
-            # The last thing to occur was an interrupt
-            # Return the value of the first interrupt as an AIMessage
-            output = langchain_to_chat_message(AIMessage(content=response["__interrupt__"][0].value))
-        else:
-            raise ValueError(f"Unexpected response type: {response_type}")
+            response_type, response = response_events[-1]
+
+            if response_type == "values" and "__interrupt__" not in response:
+                generated_message = response.get("structured_response")
+                if not generated_message:
+                    generated_message = response["messages"][-1]
+
+                # Normal response, the agent completed successfully
+                output = langchain_to_chat_message(generated_message)
+            elif response_type == "values" and "__interrupt__" in response:
+                # The last thing to occur was an interrupt
+                # Return the value of the first interrupt as an AIMessage
+                output = langchain_to_chat_message(AIMessage(content=response["__interrupt__"][0].value))
+            elif response_type == "updates" and "__interrupt__" in response:
+                # The last thing to occur was an interrupt
+                # Return the value of the first interrupt as an AIMessage
+                output = langchain_to_chat_message(AIMessage(content=response["__interrupt__"][0].value))
+            else:
+                raise ValueError(f"Unexpected response type: {response_type}")
 
-        output.run_id = str(run_id)
-        return output
+            output.run_id = str(run_id)
+            return output
 
     @handle_agent_errors
     async def stream(
@@ -391,96 +398,103 @@ class AgentExecutor:
             recursion_limit=recursion_limit,
         )
 
-        # Stream from the agent with appropriate modes
-        stream_mode = ["updates", "messages", "custom"] if stream_tokens else ["updates"]
-
-        async for stream_event in agent.graph.astream(input=input_data, config=config, stream_mode=stream_mode):
-            if not isinstance(stream_event, tuple):
-                continue
-
-            stream_mode, event = stream_event
-            new_messages = []
-
-            if stream_mode == "updates":
-                for node, updates in event.items():
-                    # A simple approach to handle agent interrupts.
-                    # In a more sophisticated implementation, we could add
-                    # some structured ChatMessage type to return the interrupt value.
-                    if node == "__interrupt__":
-                        interrupt: Interrupt
-                        for interrupt in updates:
-                            new_messages.append(AIMessage(content=interrupt.value))
-                        continue
+        # Wrap execution in trace context
+        with agent.observability.trace_context(
+            run_id=run_id,
+            user_id=user_id,
+            input=input_data,
+            agent_name=agent.name,
+        ):
+            # Stream from the agent with appropriate modes
+            stream_mode = ["updates", "messages", "custom"] if stream_tokens else ["updates"]
 
-                    update_messages = (updates or {}).get("messages", [])
-
-                    # Special case for supervisor agent
-                    if node == "supervisor":
-                        # Get only the last AIMessage since supervisor includes all previous messages
-                        ai_messages = [msg for msg in update_messages if isinstance(msg, AIMessage)]
-                        if ai_messages:
-                            update_messages = [ai_messages[-1]]
-
-                    # Special case for expert agents
-                    if node in ("research_expert", "math_expert"):
-                        # Convert to ToolMessage so it displays in the UI as a tool response
-                        if update_messages:
-                            msg = ToolMessage(
-                                content=update_messages[0].content,
-                                name=node,
-                                tool_call_id="",
-                            )
-                            update_messages = [msg]
-                    new_messages.extend(update_messages)
-
-            elif stream_mode == "custom":
-                new_messages = [event]
-
-            elif stream_mode == "messages" and stream_tokens:
-                msg, metadata = event
-                if "skip_stream" in metadata.get("tags", []):
+            async for stream_event in agent.graph.astream(input=input_data, config=config, stream_mode=stream_mode):
+                if not isinstance(stream_event, tuple):
                     continue
-                # Skip non-LLM nodes that might send messages
-                if not isinstance(msg, AIMessageChunk):
-                    continue
-                content = remove_tool_calls(msg.content)
-                if content:
-                    # Empty content in OpenAI context usually means the model is asking for a tool to be invoked
-                    yield convert_message_content_to_string(content)
-
-            # LangGraph streaming may emit tuples: (field_name, field_value)
-            # e.g. ('content', <str>), ('tool_calls', [ToolCall,...]), ('additional_kwargs', {...}), etc.
-            # We accumulate only supported fields into `parts` and skip unsupported metadata.
-            # More info at: https://langchain-ai.github.io/langgraph/cloud/how-tos/stream_messages/
-            processed_messages = []
-            current_message: dict[str, Any] = {}
-            for msg in new_messages:
-                if isinstance(msg, tuple):
-                    key, value = msg
-                    # Store parts in temporary dict
-                    current_message[key] = value
-                else:
-                    # Add complete message if we have one in progress
-                    if current_message:
-                        processed_messages.append(create_ai_message(current_message))
-                        current_message = {}
-                    processed_messages.append(msg)
-
-            # Add any remaining message parts
-            if current_message:
-                processed_messages.append(create_ai_message(current_message))
-
-            for msg in processed_messages:
-                try:
-                    chat_message = langchain_to_chat_message(msg)
-                    chat_message.run_id = str(run_id)
-                    # Skip the input message if it's repeated by LangGraph
-                    if chat_message.type == "human" and chat_message.content == msg:
+
+                stream_mode, event = stream_event
+                new_messages = []
+
+                if stream_mode == "updates":
+                    for node, updates in event.items():
+                        # A simple approach to handle agent interrupts.
+                        # In a more sophisticated implementation, we could add
+                        # some structured ChatMessage type to return the interrupt value.
+                        if node == "__interrupt__":
+                            interrupt: Interrupt
+                            for interrupt in updates:
+                                new_messages.append(AIMessage(content=interrupt.value))
+                            continue
+
+                        update_messages = (updates or {}).get("messages", [])
+
+                        # Special case for supervisor agent
+                        if node == "supervisor":
+                            # Get only the last AIMessage since supervisor includes all previous messages
+                            ai_messages = [msg for msg in update_messages if isinstance(msg, AIMessage)]
+                            if ai_messages:
+                                update_messages = [ai_messages[-1]]
+
+                        # Special case for expert agents
+                        if node in ("research_expert", "math_expert"):
+                            # Convert to ToolMessage so it displays in the UI as a tool response
+                            if update_messages:
+                                msg = ToolMessage(
+                                    content=update_messages[0].content,
+                                    name=node,
+                                    tool_call_id="",
+                                )
+                                update_messages = [msg]
+                        new_messages.extend(update_messages)
+
+                elif stream_mode == "custom":
+                    new_messages = [event]
+
+                elif stream_mode == "messages" and stream_tokens:
+                    msg, metadata = event
+                    if "skip_stream" in metadata.get("tags", []):
+                        continue
+                    # Skip non-LLM nodes that might send messages
+                    if not isinstance(msg, AIMessageChunk):
+                        continue
+                    content = remove_tool_calls(msg.content)
+                    if content:
+                        # Empty content in OpenAI context usually means the model is asking for a tool to be invoked
+                        yield convert_message_content_to_string(content)
+
+                # LangGraph streaming may emit tuples: (field_name, field_value)
+                # e.g. ('content', <str>), ('tool_calls', [ToolCall,...]), ('additional_kwargs', {...}), etc.
+                # We accumulate only supported fields into `parts` and skip unsupported metadata.
+                # More info at: https://langchain-ai.github.io/langgraph/cloud/how-tos/stream_messages/
+                processed_messages = []
+                current_message: dict[str, Any] = {}
+                for msg in new_messages:
+                    if isinstance(msg, tuple):
+                        key, value = msg
+                        # Store parts in temporary dict
+                        current_message[key] = value
+                    else:
+                        # Add complete message if we have one in progress
+                        if current_message:
+                            processed_messages.append(create_ai_message(current_message))
+                            current_message = {}
+                        processed_messages.append(msg)
+
+                # Add any remaining message parts
+                if current_message:
+                    processed_messages.append(create_ai_message(current_message))
+
+                for msg in processed_messages:
+                    try:
+                        chat_message = langchain_to_chat_message(msg)
+                        chat_message.run_id = str(run_id)
+                        # Skip the input message if it's repeated by LangGraph
+                        if chat_message.type == "human" and chat_message.content == msg:
+                            continue
+                        yield chat_message
+                    except Exception as e:
+                        logger.error(f"Error parsing message: {e}")
                         continue
-                    yield chat_message
-                except Exception as e:
-                    logger.error(f"Error parsing message: {e}")
-                    continue
 
     def save(self, path: str, agent_ids: Optional[List[str]] = None) -> None:
         """Save agents to disk using joblib.

{langgraph_agent_toolkit-0.8.0 → langgraph_agent_toolkit-0.8.2}/langgraph_agent_toolkit/agents/blueprints/bg_task_agent/agent.py

@@ -10,7 +10,7 @@ from langgraph.types import StreamWriter
 from langgraph_agent_toolkit.agents.agent import Agent
 from langgraph_agent_toolkit.agents.blueprints.bg_task_agent.task import Task
 from langgraph_agent_toolkit.core import settings
-from langgraph_agent_toolkit.core.models.factory import ModelFactory
+from langgraph_agent_toolkit.core.models.factory import CompletionModelFactory
 from langgraph_agent_toolkit.schema.models import ModelProvider
 
 
@@ -36,10 +36,10 @@ async def acall_model(state: AgentState, config: RunnableConfig) -> AgentState:
     if model_config_key and model_config_key in settings.MODEL_CONFIGS:
         # Create model from configuration
         model_config = settings.MODEL_CONFIGS[model_config_key]
-        m = ModelFactory.get_model_from_config(model_config)
+        m = CompletionModelFactory.get_model_from_config(model_config)
     else:
         # Fall back to traditional approach
-        m = ModelFactory.create(
+        m = CompletionModelFactory.create(
             mmodel_provider=config["configurable"].get("model_provider", ModelProvider.OPENAI),
             model_name=config["configurable"].get("model_name", settings.OPENAI_MODEL_NAME),
             openai_api_base=settings.OPENAI_API_BASE_URL,

{langgraph_agent_toolkit-0.8.0 → langgraph_agent_toolkit-0.8.2}/langgraph_agent_toolkit/agents/blueprints/chatbot/agent.py

@@ -4,7 +4,7 @@ from langgraph.func import entrypoint
 
 from langgraph_agent_toolkit.agents.agent import Agent
 from langgraph_agent_toolkit.core import settings
-from langgraph_agent_toolkit.core.models.factory import ModelFactory
+from langgraph_agent_toolkit.core.models.factory import CompletionModelFactory
 from langgraph_agent_toolkit.schema.models import ModelProvider
 
 
@@ -27,10 +27,10 @@ async def chatbot(
     if model_config_key and model_config_key in settings.MODEL_CONFIGS:
         # Use the model configuration from settings
         model_config = settings.MODEL_CONFIGS[model_config_key]
-        model = ModelFactory.get_model_from_config(model_config)
+        model = CompletionModelFactory.get_model_from_config(model_config)
     else:
         # Fall back to the traditional approach
-        model = ModelFactory.create(
+        model = CompletionModelFactory.create(
             model_provider=config["configurable"].get("model_provider", ModelProvider.OPENAI),
             model_name=config["configurable"].get("model_name", settings.OPENAI_MODEL_NAME),
             openai_api_base=settings.OPENAI_API_BASE_URL,

{langgraph_agent_toolkit-0.8.0 → langgraph_agent_toolkit-0.8.2}/langgraph_agent_toolkit/agents/blueprints/interrupt_agent/agent.py

@@ -13,7 +13,7 @@ from pydantic import BaseModel, Field
 
 from langgraph_agent_toolkit.agents.agent import Agent
 from langgraph_agent_toolkit.core import settings
-from langgraph_agent_toolkit.core.models.factory import ModelFactory
+from langgraph_agent_toolkit.core.models.factory import CompletionModelFactory
 from langgraph_agent_toolkit.schema.models import ModelProvider
 
 
@@ -45,7 +45,7 @@ Don't tell the user what their sign is, you are just demonstrating your knowledg
 
 async def background(state: AgentState, config: RunnableConfig) -> AgentState:
     """Demonstrate doing work before the interrupt."""
-    m = ModelFactory.create(
+    m = CompletionModelFactory.create(
         model_provider=config["configurable"].get("model_provider", ModelProvider.OPENAI),
         model_name=config["configurable"].get("model_name", settings.OPENAI_MODEL_NAME),
         openai_api_base=settings.OPENAI_API_BASE_URL,
@@ -80,7 +80,7 @@ async def determine_birthdate(state: AgentState, config: RunnableConfig) -> Agen
 
     If no birthdate is found, it will perform an interrupt before proceeding.
     """
-    m = ModelFactory.create(
+    m = CompletionModelFactory.create(
         model_provider=config["configurable"].get("model_provider", ModelProvider.OPENAI),
         model_name=config["configurable"].get("model_name", settings.OPENAI_MODEL_NAME),
         config_prefix="",
@@ -141,7 +141,7 @@ async def determine_sign(state: AgentState, config: RunnableConfig) -> AgentStat
     if not state.get("birthdate"):
         raise ValueError("No birthdate found in state")
 
-    m = ModelFactory.create(
+    m = CompletionModelFactory.create(
         model_provider=config["configurable"].get("model_provider", ModelProvider.OPENAI),
         model_name=config["configurable"].get("model_name", settings.OPENAI_MODEL_NAME),
         openai_api_base=settings.OPENAI_API_BASE_URL,

{langgraph_agent_toolkit-0.8.0 → langgraph_agent_toolkit-0.8.2}/langgraph_agent_toolkit/agents/blueprints/knowledge_base_agent/agent.py

@@ -12,7 +12,7 @@ from langgraph.managed import RemainingSteps
 
 from langgraph_agent_toolkit.agents.agent import Agent
 from langgraph_agent_toolkit.core import settings
-from langgraph_agent_toolkit.core.models.factory import ModelFactory
+from langgraph_agent_toolkit.core.models.factory import CompletionModelFactory
 from langgraph_agent_toolkit.helper.logging import logger
 from langgraph_agent_toolkit.schema.models import ModelProvider
 
@@ -158,10 +158,10 @@ async def acall_model(state: AgentState, config: RunnableConfig) -> AgentState:
     if model_config_key and model_config_key in settings.MODEL_CONFIGS:
         # Use the model configuration from settings
         model_config = settings.MODEL_CONFIGS[model_config_key]
-        model = ModelFactory.get_model_from_config(model_config)
+        model = CompletionModelFactory.get_model_from_config(model_config)
     else:
         # Fall back to the traditional approach
-        model = ModelFactory.create(
+        model = CompletionModelFactory.create(
             model_provider=config["configurable"].get("model_provider", ModelProvider.OPENAI),
             model_name=config["configurable"].get("model_name", settings.OPENAI_MODEL_NAME),
             openai_api_base=settings.OPENAI_API_BASE_URL,

{langgraph_agent_toolkit-0.8.0 → langgraph_agent_toolkit-0.8.2}/langgraph_agent_toolkit/agents/blueprints/react/agent.py

@@ -5,7 +5,7 @@ from langgraph_agent_toolkit.agents.agent import Agent
 from langgraph_agent_toolkit.agents.components.creators.create_react_agent import create_react_agent
 from langgraph_agent_toolkit.agents.components.tools import add, multiply
 from langgraph_agent_toolkit.agents.components.utils import AgentStateWithRemainingSteps, pre_model_hook_standard
-from langgraph_agent_toolkit.core.models.factory import ModelFactory
+from langgraph_agent_toolkit.core.models.factory import CompletionModelFactory
 from langgraph_agent_toolkit.core.observability.factory import ObservabilityFactory
 from langgraph_agent_toolkit.core.observability.types import ChatMessageDict, MessageRole, ObservabilityBackend
 from langgraph_agent_toolkit.core.prompts.chat_prompt_template import ObservabilityChatPromptTemplate
@@ -51,7 +51,7 @@ react_agent = Agent(
     name="react-agent",
     description="A react agent.",
     graph=create_react_agent(
-        model=ModelFactory.create(
+        model=CompletionModelFactory.create(
             model_provider=ModelProvider.OPENAI,
             model_name=settings.OPENAI_MODEL_NAME,
             openai_api_base=settings.OPENAI_API_BASE_URL,

{langgraph_agent_toolkit-0.8.0 → langgraph_agent_toolkit-0.8.2}/langgraph_agent_toolkit/agents/blueprints/react_new/agent.py

@@ -10,12 +10,12 @@ from langgraph.checkpoint.memory import MemorySaver
 from langgraph_agent_toolkit.agents.agent import Agent
 from langgraph_agent_toolkit.agents.components.tools import add, multiply
 from langgraph_agent_toolkit.core import settings
-from langgraph_agent_toolkit.core.models.factory import ModelFactory
+from langgraph_agent_toolkit.core.models.factory import CompletionModelFactory
 from langgraph_agent_toolkit.helper.constants import DEFAULT_MAX_MESSAGE_HISTORY_LENGTH
 from langgraph_agent_toolkit.schema.models import ModelProvider
 
 
-model = ModelFactory.create(
+model = CompletionModelFactory.create(
     model_provider=ModelProvider.OPENAI,
     model_name=settings.OPENAI_MODEL_NAME,
     config_prefix="",

{langgraph_agent_toolkit-0.8.0 → langgraph_agent_toolkit-0.8.2}/langgraph_agent_toolkit/agents/blueprints/react_so/agent.py

@@ -13,7 +13,7 @@ from pydantic import BaseModel, Field
 from langgraph_agent_toolkit.agents.agent import Agent
 from langgraph_agent_toolkit.agents.components.tools import add, multiply
 from langgraph_agent_toolkit.core import settings
-from langgraph_agent_toolkit.core.models.factory import ModelFactory
+from langgraph_agent_toolkit.core.models.factory import CompletionModelFactory
 from langgraph_agent_toolkit.helper.constants import DEFAULT_MAX_MESSAGE_HISTORY_LENGTH
 from langgraph_agent_toolkit.schema.models import ModelProvider
 
@@ -27,7 +27,7 @@ class ResponseSchema(BaseModel):
     )
 
 
-model = ModelFactory.create(
+model = CompletionModelFactory.create(
     model_provider=ModelProvider.OPENAI,
     model_name=settings.OPENAI_MODEL_NAME,
     config_prefix="",

{langgraph_agent_toolkit-0.8.0 → langgraph_agent_toolkit-0.8.2}/langgraph_agent_toolkit/agents/blueprints/supervisor_agent/agent.py

@@ -6,11 +6,11 @@ from langgraph_supervisor import create_supervisor
 from langgraph_agent_toolkit.agents.agent import Agent
 from langgraph_agent_toolkit.agents.components.tools import add, multiply
 from langgraph_agent_toolkit.core import settings
-from langgraph_agent_toolkit.core.models.factory import ModelFactory
+from langgraph_agent_toolkit.core.models.factory import CompletionModelFactory
 from langgraph_agent_toolkit.schema.models import ModelProvider
 
 
-model = ModelFactory.create(
+model = CompletionModelFactory.create(
     model_provider=ModelProvider.OPENAI,
     model_name=settings.OPENAI_MODEL_NAME,
     openai_api_base=settings.OPENAI_API_BASE_URL,

langgraph_agent_toolkit-0.8.2/langgraph_agent_toolkit/core/__init__.py

@@ -0,0 +1,5 @@
+from langgraph_agent_toolkit.core.models.factory import CompletionModelFactory
+from langgraph_agent_toolkit.core.settings import settings
+
+
+__all__ = ["settings", "CompletionModelFactory"]

langgraph_agent_toolkit-0.8.2/langgraph_agent_toolkit/core/models/__init__.py

@@ -0,0 +1,11 @@
+from langgraph_agent_toolkit.core.models.chat_openai import ChatOpenAIPatched
+from langgraph_agent_toolkit.core.models.factory import CompletionModelFactory, EmbeddingModelFactory
+from langgraph_agent_toolkit.core.models.fake import FakeToolModel
+
+
+__all__ = [
+    "ChatOpenAIPatched",
+    "FakeToolModel",
+    "EmbeddingModelFactory",
+    "CompletionModelFactory",
+]