hf-model (#2)

- fix: stage pre-commit (96b61e6f8ad488393a7e5d16dd3c990644ba3354)
- feat: adds hf model (341851cb943f1e90739ffdac67bacb4697ecf788)
- feat: adds langchain-huggingface (3e0cf5fcdccc19e6a99b056997b4724263338468)
- feat: removes annotations (68fd1972d30aae45b30e45e30a4797b7a3e6c4b5)
- feat: adds hf model (b86bb8539025eb94570ed91258e90c2ccecf047e)
- feat: removes annotations (23679ba810d69de9295de49c5d4e6625b3a94bd2)
- feat: updates dependencies (a16fc5edf21459e10072fce68c630d5a9b2f4df1)

.pre-commit-config.yaml

@@ -65,6 +65,7 @@ repos:
             "--format",
             "requirements-txt",
             "--no-hashes",
+            "--no-annotate",
             "--no-dev",
             "-o",
             "requirements.txt",
@@ -82,6 +83,7 @@ repos:
             "--format",
             "requirements-txt",
             "--no-hashes",
+            "--no-annotate",
             "--group",
             "dev",
             "--group",
@@ -91,7 +93,7 @@ repos:
           ]
       - id: mypy
         name: Running mypy
-        stages: [commit]
+        stages: [pre-commit]
         language: system
         entry: uv run mypy
         args: [--install-types, --non-interactive]

pyproject.toml

@@ -16,6 +16,7 @@ dependencies = [
     "gradio[mcp]~=5.31",
     "huggingface-hub>=0.32.3",
     "langchain-aws>=0.2.24",
+    "langchain-huggingface>=0.2.0",
     "langchain-mcp-adapters>=0.1.1",
     "langgraph>=0.4.7",
 ]

requirements-dev.txt

@@ -1,5 +1,5 @@
 # This file was autogenerated by uv via the following command:
-#    uv export --format requirements-txt --no-hashes --group dev --group test -o requirements-dev.txt
+#    uv export --format requirements-txt --no-hashes --no-annotate --group dev --group test -o requirements-dev.txt
 aiofiles==24.1.0
 annotated-types==0.7.0
 anyio==4.9.0
@@ -37,10 +37,12 @@ idna==3.10
 iniconfig==2.1.0
 jinja2==3.1.6
 jmespath==1.0.1
+joblib==1.5.1
 jsonpatch==1.33
 jsonpointer==3.0.0
 langchain-aws==0.2.24
 langchain-core==0.3.63
+langchain-huggingface==0.2.0
 langchain-mcp-adapters==0.1.1
 langgraph==0.4.7
 langgraph-checkpoint==2.0.26
@@ -52,12 +54,29 @@ markdown-it-py==3.0.0
 markupsafe==3.0.2
 mcp==1.9.0
 mdurl==0.1.2
+mpmath==1.3.0
 msgpack==1.1.0
 mypy==1.16.0
 mypy-extensions==1.1.0
+networkx==3.4.2 ; python_full_version < '3.11'
+networkx==3.5 ; python_full_version >= '3.11'
 nodeenv==1.9.1
 numpy==1.26.4 ; python_full_version < '3.12'
 numpy==2.2.6 ; python_full_version >= '3.12'
+nvidia-cublas-cu12==12.6.4.1 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-cuda-cupti-cu12==12.6.80 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-cuda-nvrtc-cu12==12.6.77 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-cuda-runtime-cu12==12.6.77 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-cudnn-cu12==9.5.1.17 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-cufft-cu12==11.3.0.4 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-cufile-cu12==1.11.1.6 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-curand-cu12==10.3.7.77 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-cusolver-cu12==11.7.1.2 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-cusparse-cu12==12.5.4.2 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-cusparselt-cu12==0.6.3 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-nccl-cu12==2.26.2 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-nvjitlink-cu12==12.6.85 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-nvtx-cu12==12.6.77 ; platform_machine == 'x86_64' and sys_platform == 'linux'
 orjson==3.10.18
 ormsgpack==1.10.0
 packageurl-python==0.16.0
@@ -88,24 +107,36 @@ python-dotenv==1.1.0
 python-multipart==0.0.20
 pytz==2025.2
 pyyaml==6.0.2
+regex==2024.11.6
 requests==2.32.3
 requests-toolbelt==1.0.0
 rich==14.0.0
 ruff==0.11.12
 s3transfer==0.13.0
 safehttpx==0.1.6
+safetensors==0.5.3
+scikit-learn==1.6.1
+scipy==1.15.3
 semantic-version==2.10.0
+sentence-transformers==4.1.0
+setuptools==80.9.0 ; (python_full_version >= '3.12' and platform_machine != 'x86_64') or (python_full_version >= '3.12' and sys_platform != 'linux') or (platform_machine == 'x86_64' and sys_platform == 'linux')
 shellingham==1.5.4 ; sys_platform != 'emscripten'
 six==1.17.0
 sniffio==1.3.1
 sortedcontainers==2.4.0
 sse-starlette==2.3.6
 starlette==0.46.2
+sympy==1.14.0
 tenacity==9.1.2
+threadpoolctl==3.6.0
+tokenizers==0.21.1
 toml==0.10.2
 tomli==2.2.1 ; python_full_version <= '3.11'
 tomlkit==0.13.2
+torch==2.7.1
 tqdm==4.67.1
+transformers==4.52.4
+triton==3.3.1 ; platform_machine == 'x86_64' and sys_platform == 'linux'
 typer==0.16.0 ; sys_platform != 'emscripten'
 typing-extensions==4.13.2
 typing-inspection==0.4.1

requirements.txt

@@ -1,5 +1,5 @@
 # This file was autogenerated by uv via the following command:
-#    uv export --format requirements-txt --no-hashes --no-dev -o requirements.txt
+#    uv export --format requirements-txt --no-hashes --no-annotate --no-dev -o requirements.txt
 aiofiles==24.1.0
 annotated-types==0.7.0
 anyio==4.9.0
@@ -30,10 +30,12 @@ idna==3.10
 iniconfig==2.1.0
 jinja2==3.1.6
 jmespath==1.0.1
+joblib==1.5.1
 jsonpatch==1.33
 jsonpointer==3.0.0
 langchain-aws==0.2.24
 langchain-core==0.3.63
+langchain-huggingface==0.2.0
 langchain-mcp-adapters==0.1.1
 langgraph==0.4.7
 langgraph-checkpoint==2.0.26
@@ -44,8 +46,25 @@ markdown-it-py==3.0.0 ; sys_platform != 'emscripten'
 markupsafe==3.0.2
 mcp==1.9.0
 mdurl==0.1.2 ; sys_platform != 'emscripten'
+mpmath==1.3.0
+networkx==3.4.2 ; python_full_version < '3.11'
+networkx==3.5 ; python_full_version >= '3.11'
 numpy==1.26.4 ; python_full_version < '3.12'
 numpy==2.2.6 ; python_full_version >= '3.12'
+nvidia-cublas-cu12==12.6.4.1 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-cuda-cupti-cu12==12.6.80 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-cuda-nvrtc-cu12==12.6.77 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-cuda-runtime-cu12==12.6.77 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-cudnn-cu12==9.5.1.17 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-cufft-cu12==11.3.0.4 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-cufile-cu12==1.11.1.6 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-curand-cu12==10.3.7.77 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-cusolver-cu12==11.7.1.2 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-cusparse-cu12==12.5.4.2 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-cusparselt-cu12==0.6.3 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-nccl-cu12==2.26.2 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-nvjitlink-cu12==12.6.85 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+nvidia-nvtx-cu12==12.6.77 ; platform_machine == 'x86_64' and sys_platform == 'linux'
 orjson==3.10.18
 ormsgpack==1.10.0
 packaging==24.2
@@ -66,22 +85,34 @@ python-dotenv==1.1.0
 python-multipart==0.0.20
 pytz==2025.2
 pyyaml==6.0.2
+regex==2024.11.6
 requests==2.32.3
 requests-toolbelt==1.0.0
 rich==14.0.0 ; sys_platform != 'emscripten'
 ruff==0.11.12 ; sys_platform != 'emscripten'
 s3transfer==0.13.0
 safehttpx==0.1.6
+safetensors==0.5.3
+scikit-learn==1.6.1
+scipy==1.15.3
 semantic-version==2.10.0
+sentence-transformers==4.1.0
+setuptools==80.9.0 ; (python_full_version >= '3.12' and platform_machine != 'x86_64') or (python_full_version >= '3.12' and sys_platform != 'linux') or (platform_machine == 'x86_64' and sys_platform == 'linux')
 shellingham==1.5.4 ; sys_platform != 'emscripten'
 six==1.17.0
 sniffio==1.3.1
 sse-starlette==2.3.6
 starlette==0.46.2
+sympy==1.14.0
 tenacity==9.1.2
+threadpoolctl==3.6.0
+tokenizers==0.21.1
 tomli==2.2.1 ; python_full_version <= '3.11'
 tomlkit==0.13.2
+torch==2.7.1
 tqdm==4.67.1
+transformers==4.52.4
+triton==3.3.1 ; platform_machine == 'x86_64' and sys_platform == 'linux'
 typer==0.16.0 ; sys_platform != 'emscripten'
 typing-extensions==4.13.2
 typing-inspection==0.4.1

tdagent/grchat.py

@@ -10,6 +10,7 @@ import botocore.exceptions
 import gradio as gr
 from langchain_aws import ChatBedrock
 from langchain_core.messages import AIMessage, HumanMessage, SystemMessage
+from langchain_huggingface import HuggingFaceEndpoint
 from langchain_mcp_adapters.client import MultiServerMCPClient
 from langgraph.prebuilt import create_react_agent
 
@@ -56,6 +57,7 @@ llm_agent: CompiledGraph | None = None
 #### Utility functions ####
 
 
+## Bedrock LLM creation ##
 def create_bedrock_llm(
     bedrock_model_id: str,
     aws_access_key: str,
@@ -91,9 +93,25 @@ def create_bedrock_llm(
     return llm, ""
 
 
-#### UI functionality ####
+## Hugging Face LLM creation ##
+def create_hf_llm(
+    hf_model_id: str,
+    huggingfacehub_api_token: str | None = None,
+) -> tuple[HuggingFaceEndpoint | None, str]:
+    """Create a LangGraph Hugging Face agent."""
+    try:
+        llm = HuggingFaceEndpoint(
+            model=hf_model_id,
+            huggingfacehub_api_token=huggingfacehub_api_token,
+            temperature=0.8,
+        )
+    except Exception as e:  # noqa: BLE001
+        return None, str(e)
+
+    return llm, ""
 
 
+#### UI functionality ####
 async def gr_connect_to_bedrock(
     model_id: str,
     access_key: str,
@@ -128,7 +146,7 @@ async def gr_connect_to_bedrock(
     #     }
     # )
     # tools = await client.get_tools()
-
+    tools = []
     if mcp_servers:
         client = MultiServerMCPClient(
             {
@@ -140,8 +158,6 @@ async def gr_connect_to_bedrock(
             },
         )
         tools = await client.get_tools()
-    else:
-        tools = []
 
     llm_agent = create_react_agent(
         model=llm,
@@ -152,6 +168,40 @@ async def gr_connect_to_bedrock(
     return "✅ Successfully connected to AWS Bedrock!"
 
 
+async def gr_connect_to_hf(
+    model_id: str,
+    hf_access_token_textbox: str | None,
+    mcp_servers: Sequence[MutableCheckBoxGroupEntry] | None,
+) -> str:
+    """Initialize Hugging Face agent."""
+    global llm_agent  # noqa: PLW0603
+
+    llm, error = create_hf_llm(model_id, hf_access_token_textbox)
+
+    if llm is None:
+        return f"❌ Connection failed: {error}"
+    tools = []
+    if mcp_servers:
+        client = MultiServerMCPClient(
+            {
+                server.name.replace(" ", "-"): {
+                    "url": server.value,
+                    "transport": "sse",
+                }
+                for server in mcp_servers
+            },
+        )
+        tools = await client.get_tools()
+
+    llm_agent = create_react_agent(
+        model=llm,
+        tools=tools,
+        prompt=SYSTEM_MESSAGE,
+    )
+
+    return "✅ Successfully connected to Hugging Face!"
+
+
 async def gr_chat_function(  # noqa: D103
     message: str,
     history: list[Mapping[str, str]],
@@ -249,6 +299,32 @@ with gr.Blocks() as gr_app:
             outputs=[status_textbox],
         )
 
+    with gr.Accordion("Hugging Face Configuration", open=True):
+        with gr.Row():
+            hf_model_id_textbox = gr.Textbox(
+                label="HF Model Id",
+                value="fdtn-ai/Foundation-Sec-8B",
+            )
+        with gr.Row():
+            hf_access_token_textbox = gr.Textbox(
+                label="Hugging Face Access Token",
+                type="password",
+                placeholder="Enter your Hugging Face Access Token",
+            )
+            hf_connect_btn = gr.Button("🔌 Connect to Hugging Face", variant="primary")
+
+        status_textbox = gr.Textbox(label="Connection Status", interactive=False)
+
+        hf_connect_btn.click(
+            gr_connect_to_hf,
+            inputs=[
+                hf_model_id_textbox,
+                hf_access_token_textbox,
+                mcp_list.state,
+            ],
+            outputs=[status_textbox],
+        )
+
     chat_interface = gr.ChatInterface(
         fn=gr_chat_function,
         type="messages",