Make TGI adapter compatible with HF Inference API (#97)

llama_stack/distribution/templates/local-hf-endpoint-build.yaml

@@ -0,0 +1,10 @@
+name: local-hf-endpoint
+distribution_spec:
+  description: "Like local, but use Hugging Face Inference Endpoints for running LLM inference.\nSee https://hf.co/docs/api-endpoints."
+  providers:
+    inference: remote::hf::endpoint
+    memory: meta-reference
+    safety: meta-reference
+    agents: meta-reference
+    telemetry: meta-reference
+image_type: conda

llama_stack/distribution/templates/local-hf-serverless-build.yaml

@@ -0,0 +1,10 @@
+name: local-hf-serverless
+distribution_spec:
+  description: "Like local, but use Hugging Face Inference API (serverless) for running LLM inference.\nSee https://hf.co/docs/api-inference."
+  providers:
+    inference: remote::hf::serverless
+    memory: meta-reference
+    safety: meta-reference
+    agents: meta-reference
+    telemetry: meta-reference
+image_type: conda

llama_stack/distribution/templates/local-tgi-build.yaml

@@ -1,6 +1,6 @@
 name: local-tgi
 distribution_spec:
-  description: Use TGI (local or with Hugging Face Inference Endpoints for running LLM inference. When using HF Inference Endpoints, you must provide the name of the endpoint).
+  description: Like local, but use a TGI server for running LLM inference.
   providers:
     inference: remote::tgi
     memory: meta-reference

llama_stack/providers/adapters/inference/tgi/__init__.py

@@ -4,21 +4,26 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from .config import TGIImplConfig
+from typing import Union
-from .tgi import InferenceEndpointAdapter, TGIAdapter
+
+from .config import InferenceAPIImplConfig, InferenceEndpointImplConfig, TGIImplConfig
+from .tgi import InferenceAPIAdapter, InferenceEndpointAdapter, TGIAdapter
 
 
-async def get_adapter_impl(config: TGIImplConfig, _deps):
+async def get_adapter_impl(
-    assert isinstance(config, TGIImplConfig), f"Unexpected config type: {type(config)}"
+    config: Union[InferenceAPIImplConfig, InferenceEndpointImplConfig, TGIImplConfig],
-
+    _deps,
-    if config.url is not None:
+):
-        impl = TGIAdapter(config)
+    if isinstance(config, TGIImplConfig):
-    elif config.is_inference_endpoint():
+        impl = TGIAdapter()
-        impl = InferenceEndpointAdapter(config)
+    elif isinstance(config, InferenceAPIImplConfig):
+        impl = InferenceAPIAdapter()
+    elif isinstance(config, InferenceEndpointImplConfig):
+        impl = InferenceEndpointAdapter()
     else:
         raise ValueError(
-            "Invalid configuration. Specify either an URL or HF Inference Endpoint details (namespace and endpoint name)."
+            f"Invalid configuration. Expected 'TGIAdapter', 'InferenceAPIImplConfig' or 'InferenceEndpointImplConfig'. Got {type(config)}."
         )
 
-    await impl.initialize()
+    await impl.initialize(config)
     return impl

llama_stack/providers/adapters/inference/tgi/config.py

@@ -12,18 +12,32 @@ from pydantic import BaseModel, Field
 
 @json_schema_type
 class TGIImplConfig(BaseModel):
-    url: Optional[str] = Field(
+    url: str = Field(
-        default=None,
+        description="The URL for the TGI endpoint (e.g. 'http://localhost:8080')",
-        description="The URL for the local TGI endpoint (e.g., http://localhost:8080)",
     )
     api_token: Optional[str] = Field(
         default=None,
-        description="The HF token for Hugging Face Inference Endpoints (will default to locally saved token if not provided)",
+        description="A bearer token if your TGI endpoint is protected.",
-    )
-    hf_endpoint_name: Optional[str] = Field(
-        default=None,
-        description="The name of the Hugging Face Inference Endpoint : can be either in the format of '{namespace}/{endpoint_name}' (namespace can be the username or organization name) or just '{endpoint_name}' if logged into the same account as the namespace",
     )
 
-    def is_inference_endpoint(self) -> bool:
+
-        return self.hf_endpoint_name is not None
+@json_schema_type
+class InferenceEndpointImplConfig(BaseModel):
+    endpoint_name: str = Field(
+        description="The name of the Hugging Face Inference Endpoint in the format of '{namespace}/{endpoint_name}' (e.g. 'my-cool-org/meta-llama-3-1-8b-instruct-rce'). Namespace is optional and will default to the user account if not provided.",
+    )
+    api_token: Optional[str] = Field(
+        default=None,
+        description="Your Hugging Face user access token (will default to locally saved token if not provided)",
+    )
+
+
+@json_schema_type
+class InferenceAPIImplConfig(BaseModel):
+    model_id: str = Field(
+        description="The model ID of the model on the Hugging Face Hub (e.g. 'meta-llama/Meta-Llama-3.1-70B-Instruct')",
+    )
+    api_token: Optional[str] = Field(
+        default=None,
+        description="Your Hugging Face user access token (will default to locally saved token if not provided)",
+    )

llama_stack/providers/adapters/inference/tgi/tgi.py

@@ -5,54 +5,33 @@
 # the root directory of this source tree.
 
 
-from typing import Any, AsyncGenerator, Dict
+import logging
+from typing import AsyncGenerator
 
-import requests
+from huggingface_hub import AsyncInferenceClient, HfApi
-
-from huggingface_hub import HfApi, InferenceClient
 from llama_models.llama3.api.chat_format import ChatFormat
 from llama_models.llama3.api.datatypes import StopReason
 from llama_models.llama3.api.tokenizer import Tokenizer
+
 from llama_stack.apis.inference import *  # noqa: F403
 from llama_stack.providers.utils.inference.augment_messages import (
     augment_messages_for_tools,
 )
 
-from .config import TGIImplConfig
+from .config import InferenceAPIImplConfig, InferenceEndpointImplConfig, TGIImplConfig
+
+logger = logging.getLogger(__name__)
 
 
-class TGIAdapter(Inference):
+class _HfAdapter(Inference):
-    def __init__(self, config: TGIImplConfig) -> None:
+    client: AsyncInferenceClient
-        self.config = config
+    max_tokens: int
+    model_id: str
+
+    def __init__(self) -> None:
         self.tokenizer = Tokenizer.get_instance()
         self.formatter = ChatFormat(self.tokenizer)
 
-    @property
-    def client(self) -> InferenceClient:
-        return InferenceClient(model=self.config.url, token=self.config.api_token)
-
-    def _get_endpoint_info(self) -> Dict[str, Any]:
-        return {
-            **self.client.get_endpoint_info(),
-            "inference_url": self.config.url,
-        }
-
-    async def initialize(self) -> None:
-        try:
-            info = self._get_endpoint_info()
-            if "model_id" not in info:
-                raise RuntimeError("Missing model_id in model info")
-            if "max_total_tokens" not in info:
-                raise RuntimeError("Missing max_total_tokens in model info")
-            self.max_tokens = info["max_total_tokens"]
-
-            self.inference_url = info["inference_url"]
-        except Exception as e:
-            import traceback
-
-            traceback.print_exc()
-            raise RuntimeError(f"Error initializing TGIAdapter: {e}") from e
-
     async def shutdown(self) -> None:
         pass
 
@@ -111,7 +90,7 @@ class TGIAdapter(Inference):
 
         options = self.get_chat_options(request)
         if not request.stream:
-            response = self.client.text_generation(
+            response = await self.client.text_generation(
                 prompt=prompt,
                 stream=False,
                 details=True,
@@ -147,7 +126,7 @@ class TGIAdapter(Inference):
             stop_reason = None
             tokens = []
 
-            for response in self.client.text_generation(
+            async for response in await self.client.text_generation(
                 prompt=prompt,
                 stream=True,
                 details=True,
@@ -239,46 +218,36 @@ class TGIAdapter(Inference):
             )
 
 
-class InferenceEndpointAdapter(TGIAdapter):
+class TGIAdapter(_HfAdapter):
-    def __init__(self, config: TGIImplConfig) -> None:
+    async def initialize(self, config: TGIImplConfig) -> None:
-        super().__init__(config)
+        self.client = AsyncInferenceClient(model=config.url, token=config.api_token)
-        self.config.url = self._construct_endpoint_url()
+        endpoint_info = await self.client.get_endpoint_info()
+        self.max_tokens = endpoint_info["max_total_tokens"]
+        self.model_id = endpoint_info["model_id"]
 
-    def _construct_endpoint_url(self) -> str:
+
-        hf_endpoint_name = self.config.hf_endpoint_name
+class InferenceAPIAdapter(_HfAdapter):
-        assert hf_endpoint_name.count("/") <= 1, (
+    async def initialize(self, config: InferenceAPIImplConfig) -> None:
-            "Endpoint name must be in the format of 'namespace/endpoint_name' "
+        self.client = AsyncInferenceClient(
-            "or 'endpoint_name'"
+            model=config.model_id, token=config.api_token
         )
-        if "/" not in hf_endpoint_name:
+        endpoint_info = await self.client.get_endpoint_info()
-            hf_namespace: str = self.get_namespace()
+        self.max_tokens = endpoint_info["max_total_tokens"]
-            endpoint_path = f"{hf_namespace}/{hf_endpoint_name}"
+        self.model_id = endpoint_info["model_id"]
-        else:
-            endpoint_path = hf_endpoint_name
-        return f"https://api.endpoints.huggingface.cloud/v2/endpoint/{endpoint_path}"
 
-    def get_namespace(self) -> str:
-        return HfApi().whoami()["name"]
 
-    @property
+class InferenceEndpointAdapter(_HfAdapter):
-    def client(self) -> InferenceClient:
+    async def initialize(self, config: InferenceEndpointImplConfig) -> None:
-        return InferenceClient(model=self.inference_url, token=self.config.api_token)
+        # Get the inference endpoint details
+        api = HfApi(token=config.api_token)
+        endpoint = api.get_inference_endpoint(config.endpoint_name)
 
-    def _get_endpoint_info(self) -> Dict[str, Any]:
+        # Wait for the endpoint to be ready (if not already)
-        headers = {
+        endpoint.wait(timeout=60)
-            "accept": "application/json",
-            "authorization": f"Bearer {self.config.api_token}",
-        }
-        response = requests.get(self.config.url, headers=headers)
-        response.raise_for_status()
-        endpoint_info = response.json()
-        return {
-            "inference_url": endpoint_info["status"]["url"],
-            "model_id": endpoint_info["model"]["repository"],
-            "max_total_tokens": int(
-                endpoint_info["model"]["image"]["custom"]["env"]["MAX_TOTAL_TOKENS"]
-            ),
-        }
 
-    async def initialize(self) -> None:
+        # Initialize the adapter
-        await super().initialize()
+        self.client = endpoint.async_client
+        self.model_id = endpoint.repository
+        self.max_tokens = int(
+            endpoint.raw["model"]["image"]["custom"]["env"]["MAX_TOTAL_TOKENS"]
+        )

llama_stack/providers/registry/inference.py

@@ -48,11 +48,29 @@ def available_providers() -> List[ProviderSpec]:
             api=Api.inference,
             adapter=AdapterSpec(
                 adapter_id="tgi",
-                pip_packages=["huggingface_hub"],
+                pip_packages=["huggingface_hub", "aiohttp"],
                 module="llama_stack.providers.adapters.inference.tgi",
                 config_class="llama_stack.providers.adapters.inference.tgi.TGIImplConfig",
             ),
         ),
+        remote_provider_spec(
+            api=Api.inference,
+            adapter=AdapterSpec(
+                adapter_id="hf::serverless",
+                pip_packages=["huggingface_hub", "aiohttp"],
+                module="llama_stack.providers.adapters.inference.tgi",
+                config_class="llama_stack.providers.adapters.inference.tgi.InferenceAPIImplConfig",
+            ),
+        ),
+        remote_provider_spec(
+            api=Api.inference,
+            adapter=AdapterSpec(
+                adapter_id="hf::endpoint",
+                pip_packages=["huggingface_hub", "aiohttp"],
+                module="llama_stack.providers.adapters.inference.tgi",
+                config_class="llama_stack.providers.adapters.inference.tgi.InferenceEndpointImplConfig",
+            ),
+        ),
         remote_provider_spec(
             api=Api.inference,
             adapter=AdapterSpec(