Litellm support in llama stack:

llama_stack/providers/registry/inference.py

@@ -164,6 +164,15 @@ def available_providers() -> List[ProviderSpec]:
                 provider_data_validator="llama_stack.providers.remote.inference.groq.GroqProviderDataValidator",
             ),
         ),
+        remote_provider_spec(
+            api=Api.inference,
+            adapter=AdapterSpec(
+                adapter_type="litellm",
+                pip_packages=["litellm"],
+                module="llama_stack.providers.remote.inference.litellm",
+                config_class="llama_stack.providers.remote.inference.litellm.LitellmConfig",
+            ),
+        ),
         remote_provider_spec(
             api=Api.inference,
             adapter=AdapterSpec(

llama_stack/providers/remote/inference/litellm/__init__.py

@@ -0,0 +1,19 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from pydantic import BaseModel
+
+from llama_stack.apis.inference import Inference
+
+from .config import LitellmConfig
+
+
+async def get_adapter_impl(config: LitellmConfig, _deps) -> Inference:
+    # import dynamically so the import is used only when it is needed
+    from .litellm import LitellmInferenceAdapter
+    assert isinstance(config, LitellmConfig), f"Unexpected config type: {type(config)}"
+    adapter = LitellmInferenceAdapter(config)
+    return adapter

llama_stack/providers/remote/inference/litellm/config.py

@@ -0,0 +1,19 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+from typing import Optional
+
+from pydantic import BaseModel, Field
+
+
+class LitellmConfig(BaseModel):
+    openai_api_key: Optional[str] = Field(
+        default=None,
+        description="The access key to use for openai. Default use environment variable: OPENAI_API_KEY",
+    )
+    llm_provider: Optional[str] = Field(
+        default="openai",
+        description="The provider to use. Default use environment variable: LLM_PROVIDER",
+    )

llama_stack/providers/remote/inference/litellm/litellm.py

@@ -0,0 +1,116 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+import os
+from typing import AsyncIterator, List, Optional, Union, Any
+
+from litellm import completion as litellm_completion
+from litellm.types.utils import ModelResponse
+
+from llama_models.datatypes import SamplingParams
+from llama_models.llama3.api.datatypes import ToolDefinition, ToolPromptFormat, StopReason
+from llama_stack.apis.inference import (
+    ChatCompletionResponse,
+    ChatCompletionResponseStreamChunk,
+    CompletionResponse,
+    CompletionMessage,
+    CompletionResponseStreamChunk,
+    EmbeddingsResponse,
+    Inference,
+    InterleavedContent,
+    LogProbConfig,
+    Message,
+    ResponseFormat,
+    ToolChoice,
+)
+# from llama_stack.distribution.request_headers import NeedsRequestProviderData
+from llama_stack.providers.remote.inference.litellm.config import LitellmConfig
+from llama_stack.providers.utils.inference.model_registry import (
+    build_model_alias,
+    ModelRegistryHelper,
+)
+
+_MODEL_ALIASES = [
+    build_model_alias(
+        "gpt-4o",  # provider_model_id
+        "gpt-4o",  # model_descriptor
+    ),
+]
+
+class LitellmInferenceAdapter(Inference, ModelRegistryHelper):
+    _config: LitellmConfig
+
+    def __init__(self, config: LitellmConfig):
+        ModelRegistryHelper.__init__(self, model_aliases=_MODEL_ALIASES)
+        self._config = config
+
+    def completion(
+        self,
+        model_id: str,
+        content: InterleavedContent,
+        sampling_params: Optional[SamplingParams] = SamplingParams(),
+        response_format: Optional[ResponseFormat] = None,
+        stream: Optional[bool] = False,
+        logprobs: Optional[LogProbConfig] = None,
+    ) -> Union[CompletionResponse, AsyncIterator[CompletionResponseStreamChunk]]:
+        # litellm doesn't support non-chat completion as of time of writing
+        raise NotImplementedError()
+
+    def _messages_to_litellm_messages(
+        self,
+        messages: List[Message],
+    ) -> list[dict[str, Any]]:
+        litellm_messages = []
+        for message in messages:
+            lm_message = {
+                "role": message.role,
+                "content": message.content,
+            }
+            litellm_messages.append(lm_message)
+        return litellm_messages
+
+    def _convert_to_llama_stack_response(
+        self,
+        litellm_response: ModelResponse,
+    ) -> ChatCompletionResponse:
+        assert litellm_response.choices is not None
+        assert len(litellm_response.choices) == 1
+        message = litellm_response.choices[0].message
+        completion_message = CompletionMessage(content=message["content"], role=message["role"], stop_reason=StopReason.end_of_message, tool_calls=[])
+        return ChatCompletionResponse(completion_message=completion_message)
+
+    async def chat_completion(
+        self,
+        model_id: str,
+        messages: List[Message],
+        sampling_params: Optional[SamplingParams] = SamplingParams(),
+        response_format: Optional[ResponseFormat] = None,
+        tools: Optional[List[ToolDefinition]] = None,
+        tool_choice: Optional[ToolChoice] = ToolChoice.auto,
+        tool_prompt_format: Optional[ToolPromptFormat] = None,
+        stream: Optional[bool] = False,
+        logprobs: Optional[LogProbConfig] = None,
+    ) -> Union[
+        ChatCompletionResponse, AsyncIterator[ChatCompletionResponseStreamChunk]
+    ]:
+        assert stream is False, "streaming not supported"
+        model_id = self.get_provider_model_id(model_id)
+        response = litellm_completion(
+            model=model_id,
+            custom_llm_provider=self._config.llm_provider,
+            messages=self._messages_to_litellm_messages(messages),
+            api_key=self._config.openai_api_key,
+        )
+
+        return self._convert_to_llama_stack_response(response)
+
+    async def embeddings(
+        self,
+        model_id: str,
+        contents: List[InterleavedContent],
+    ) -> EmbeddingsResponse:
+        raise NotImplementedError()
+        

llama_stack_server-run.yaml

@@ -0,0 +1,79 @@
+version: '2'
+image_name: llama_stack_server
+container_image: null
+apis:
+- inference
+- safety
+- agents
+- vector_io
+- datasetio
+- scoring
+- eval
+- post_training
+- tool_runtime
+- telemetry
+providers:
+  inference:
+  - provider_id: litellm
+    provider_type: remote::litellm
+    config:
+      openai_api_key: ???
+  safety:
+  - provider_id: llama-guard
+    provider_type: inline::llama-guard
+    config: {}
+  agents:
+  - provider_id: meta-reference
+    provider_type: inline::meta-reference
+    config:
+      persistence_store:
+        type: sqlite
+        namespace: null
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/llama_stack_server}/agents_store.db
+  vector_io:
+  - provider_id: faiss
+    provider_type: inline::faiss
+    config:
+      kvstore:
+        type: sqlite
+        namespace: null
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/llama_stack_server}/faiss_store.db
+  datasetio:
+  - provider_id: localfs
+    provider_type: inline::localfs
+    config: {}
+  scoring:
+  - provider_id: basic
+    provider_type: inline::basic
+    config: {}
+  eval:
+  - provider_id: meta-reference
+    provider_type: inline::meta-reference
+    config: {}
+  post_training:
+  - provider_id: torchtune
+    provider_type: inline::torchtune
+    config: {}
+  tool_runtime:
+  - provider_id: rag-runtime
+    provider_type: inline::rag-runtime
+    config: {}
+  telemetry:
+  - provider_id: meta-reference
+    provider_type: inline::meta-reference
+    config:
+      service_name: ${env.OTEL_SERVICE_NAME:llama-stack}
+      sinks: ${env.TELEMETRY_SINKS:console,sqlite}
+      sqlite_db_path: ${env.SQLITE_DB_PATH:~/.llama/distributions/llama_stack_server/trace_store.db}
+metadata_store: null
+models:
+- metadata: {}
+  model_id: gpt-4o
+  provider_id: litellm
+  model_type: llm
+shields: []
+vector_dbs: []
+datasets: []
+scoring_fns: []
+eval_tasks: []
+tool_groups: []

meta-reference-gpu-run.yaml

@@ -0,0 +1,117 @@
+version: '2'
+image_name: meta-reference-gpu
+apis:
+- agents
+- datasetio
+- eval
+- inference
+- safety
+- scoring
+- telemetry
+- tool_runtime
+- vector_io
+providers:
+  inference:
+  - provider_id: meta-reference-inference
+    provider_type: inline::meta-reference
+    config:
+      model: Llama3.3-70B-Instruct
+      max_seq_len: 64000
+      checkpoint_dir: ${env.INFERENCE_CHECKPOINT_DIR:null}
+  - provider_id: sentence-transformers
+    provider_type: inline::sentence-transformers
+    config: {}
+  vector_io:
+  - provider_id: faiss
+    provider_type: inline::faiss
+    config:
+      kvstore:
+        type: sqlite
+        namespace: null
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/meta-reference-gpu}/faiss_store.db
+  safety:
+  - provider_id: llama-guard
+    provider_type: inline::llama-guard
+    config: {}
+  agents:
+  - provider_id: meta-reference
+    provider_type: inline::meta-reference
+    config:
+      persistence_store:
+        type: sqlite
+        namespace: null
+        db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/meta-reference-gpu}/agents_store.db
+  telemetry:
+  - provider_id: meta-reference
+    provider_type: inline::meta-reference
+    config:
+      service_name: ${env.OTEL_SERVICE_NAME:llama-stack}
+      sinks: ${env.TELEMETRY_SINKS:console,sqlite}
+      sqlite_db_path: ${env.SQLITE_DB_PATH:~/.llama/distributions/meta-reference-gpu/trace_store.db}
+  eval:
+  - provider_id: meta-reference
+    provider_type: inline::meta-reference
+    config: {}
+  datasetio:
+  - provider_id: huggingface
+    provider_type: remote::huggingface
+    config: {}
+  - provider_id: localfs
+    provider_type: inline::localfs
+    config: {}
+  scoring:
+  - provider_id: basic
+    provider_type: inline::basic
+    config: {}
+  - provider_id: llm-as-judge
+    provider_type: inline::llm-as-judge
+    config: {}
+  - provider_id: braintrust
+    provider_type: inline::braintrust
+    config:
+      openai_api_key: ${env.OPENAI_API_KEY:}
+  tool_runtime:
+  - provider_id: brave-search
+    provider_type: remote::brave-search
+    config:
+      api_key: ${env.BRAVE_SEARCH_API_KEY:}
+      max_results: 3
+  - provider_id: tavily-search
+    provider_type: remote::tavily-search
+    config:
+      api_key: ${env.TAVILY_SEARCH_API_KEY:}
+      max_results: 3
+  - provider_id: code-interpreter
+    provider_type: inline::code-interpreter
+    config: {}
+  - provider_id: rag-runtime
+    provider_type: inline::rag-runtime
+    config: {}
+  - provider_id: model-context-protocol
+    provider_type: remote::model-context-protocol
+    config: {}
+metadata_store:
+  type: sqlite
+  db_path: ${env.SQLITE_STORE_DIR:~/.llama/distributions/meta-reference-gpu}/registry.db
+models:
+- metadata: {}
+  model_id: Llama3.3-70B-Instruct
+  provider_id: meta-reference-inference
+  model_type: llm
+- metadata:
+    embedding_dimension: 384
+  model_id: all-MiniLM-L6-v2
+  provider_id: sentence-transformers
+  model_type: embedding
+shields: []
+vector_dbs: []
+datasets: []
+scoring_fns: []
+eval_tasks: []
+tool_groups:
+- toolgroup_id: builtin::websearch
+  provider_id: tavily-search
+- toolgroup_id: builtin::rag
+  provider_id: rag-runtime
+- toolgroup_id: builtin::code_interpreter
+  provider_id: code-interpreter