apis, alt

# What does this PR do?


## Test Plan
# What does this PR do?


## Test Plan

llama_stack/providers/inline/inference/meta_reference/inference.py

@@ -28,7 +28,7 @@ from llama_stack.apis.inference import (
     CompletionRequest,
     CompletionResponse,
     CompletionResponseStreamChunk,
-    Inference,
+    InferenceProvider,
     InterleavedContent,
     LogProbConfig,
     Message,
@@ -86,7 +86,7 @@ class MetaReferenceInferenceImpl(
     OpenAICompletionToLlamaStackMixin,
     OpenAIChatCompletionToLlamaStackMixin,
     SentenceTransformerEmbeddingMixin,
-    Inference,
+    InferenceProvider,
     ModelsProtocolPrivate,
 ):
     def __init__(self, config: MetaReferenceInferenceConfig) -> None:

llama_stack/providers/inline/inference/sentence_transformers/sentence_transformers.py

@@ -9,7 +9,7 @@ from collections.abc import AsyncGenerator
 
 from llama_stack.apis.inference import (
     CompletionResponse,
-    Inference,
+    InferenceProvider,
     InterleavedContent,
     LogProbConfig,
     Message,
@@ -38,7 +38,7 @@ class SentenceTransformersInferenceImpl(
     OpenAIChatCompletionToLlamaStackMixin,
     OpenAICompletionToLlamaStackMixin,
     SentenceTransformerEmbeddingMixin,
-    Inference,
+    InferenceProvider,
     ModelsProtocolPrivate,
 ):
     def __init__(self, config: SentenceTransformersInferenceConfig) -> None:

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

@@ -4,12 +4,12 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack.apis.inference import Inference
+from llama_stack.apis.inference import InferenceProvider
 
 from .config import CerebrasCompatConfig
 
 
-async def get_adapter_impl(config: CerebrasCompatConfig, _deps) -> Inference:
+async def get_adapter_impl(config: CerebrasCompatConfig, _deps) -> InferenceProvider:
     # import dynamically so the import is used only when it is needed
     from .cerebras import CerebrasCompatInferenceAdapter
 

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

@@ -4,12 +4,12 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack.apis.inference import Inference
+from llama_stack.apis.inference import InferenceProvider
 
 from .config import FireworksCompatConfig
 
 
-async def get_adapter_impl(config: FireworksCompatConfig, _deps) -> Inference:
+async def get_adapter_impl(config: FireworksCompatConfig, _deps) -> InferenceProvider:
     # import dynamically so the import is used only when it is needed
     from .fireworks import FireworksCompatInferenceAdapter
 

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

@@ -4,12 +4,12 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack.apis.inference import Inference
+from llama_stack.apis.inference import InferenceProvider
 
 from .config import GroqCompatConfig
 
 
-async def get_adapter_impl(config: GroqCompatConfig, _deps) -> Inference:
+async def get_adapter_impl(config: GroqCompatConfig, _deps) -> InferenceProvider:
     # import dynamically so the import is used only when it is needed
     from .groq import GroqCompatInferenceAdapter
 

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

@@ -4,12 +4,12 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack.apis.inference import Inference
+from llama_stack.apis.inference import InferenceProvider
 
 from .config import LlamaCompatConfig
 
 
-async def get_adapter_impl(config: LlamaCompatConfig, _deps) -> Inference:
+async def get_adapter_impl(config: LlamaCompatConfig, _deps) -> InferenceProvider:
     # import dynamically so the import is used only when it is needed
     from .llama import LlamaCompatInferenceAdapter
 

llama_stack/providers/remote/inference/ollama/ollama.py

@@ -28,7 +28,7 @@ from llama_stack.apis.inference import (
     EmbeddingsResponse,
     EmbeddingTaskType,
     GrammarResponseFormat,
-    Inference,
+    InferenceProvider,
     JsonSchemaResponseFormat,
     LogProbConfig,
     Message,
@@ -82,7 +82,7 @@ logger = get_logger(name=__name__, category="inference")
 
 
 class OllamaInferenceAdapter(
-    Inference,
+    InferenceProvider,
     ModelsProtocolPrivate,
 ):
     def __init__(self, url: str) -> None:

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

@@ -4,12 +4,12 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack.apis.inference import Inference
+from llama_stack.apis.inference import InferenceProvider
 
 from .config import SambaNovaCompatConfig
 
 
-async def get_adapter_impl(config: SambaNovaCompatConfig, _deps) -> Inference:
+async def get_adapter_impl(config: SambaNovaCompatConfig, _deps) -> InferenceProvider:
     # import dynamically so the import is used only when it is needed
     from .sambanova import SambaNovaCompatInferenceAdapter
 

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

@@ -4,12 +4,12 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-from llama_stack.apis.inference import Inference
+from llama_stack.apis.inference import InferenceProvider
 
 from .config import TogetherCompatConfig
 
 
-async def get_adapter_impl(config: TogetherCompatConfig, _deps) -> Inference:
+async def get_adapter_impl(config: TogetherCompatConfig, _deps) -> InferenceProvider:
     # import dynamically so the import is used only when it is needed
     from .together import TogetherCompatInferenceAdapter
 

llama_stack/providers/utils/inference/litellm_openai_mixin.py

@@ -19,7 +19,7 @@ from llama_stack.apis.inference import (
     ChatCompletionResponseStreamChunk,
     EmbeddingsResponse,
     EmbeddingTaskType,
-    Inference,
+    InferenceProvider,
     JsonSchemaResponseFormat,
     LogProbConfig,
     Message,
@@ -59,7 +59,7 @@ logger = get_logger(name=__name__, category="inference")
 
 class LiteLLMOpenAIMixin(
     ModelRegistryHelper,
-    Inference,
+    InferenceProvider,
     NeedsRequestProviderData,
 ):
     # TODO: avoid exposing the litellm specific model names to the user.