phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-08-06 18:50:44 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

add async to get_model signature in Protocol

Browse source 
Signed-off-by: Ihar Hrachyshka <ihar.hrachyshka@gmail.com>
This commit is contained in:
Ihar Hrachyshka 2025-03-25 18:25:58 -04:00
parent 6aedfc2201
commit fcf3b0a835
3 changed files with 7 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

2
llama_stack/apis/inference/inference.py
View file

@ -394,7 +394,7 @@ class EmbeddingsResponse(BaseModel):
class ModelStore(Protocol): class ModelStore(Protocol):
def get_model(self, identifier: str) -> Model: ... async def get_model(self, identifier: str) -> Model: ...
class TextTruncation(Enum): class TextTruncation(Enum):

6
llama_stack/providers/remote/inference/ollama/ollama.py
View file

@ -103,7 +103,7 @@ class OllamaInferenceAdapter(Inference, ModelsProtocolPrivate):
assert self.model_store is not None assert self.model_store is not None
if sampling_params is None: if sampling_params is None:
sampling_params = SamplingParams() sampling_params = SamplingParams()
model = self.model_store.get_model(model_id) model = await self.model_store.get_model(model_id)
request = CompletionRequest( request = CompletionRequest(
model=model.provider_resource_id, model=model.provider_resource_id,
content=content, content=content,
@ -167,7 +167,7 @@ class OllamaInferenceAdapter(Inference, ModelsProtocolPrivate):
assert self.model_store is not None assert self.model_store is not None
if sampling_params is None: if sampling_params is None:
sampling_params = SamplingParams() sampling_params = SamplingParams()
model = self.model_store.get_model(model_id) model = await self.model_store.get_model(model_id)
request = ChatCompletionRequest( request = ChatCompletionRequest(
model=model.provider_resource_id, model=model.provider_resource_id,
messages=messages, messages=messages,
@ -287,7 +287,7 @@ class OllamaInferenceAdapter(Inference, ModelsProtocolPrivate):
task_type: Optional[EmbeddingTaskType] = None, task_type: Optional[EmbeddingTaskType] = None,
) -> EmbeddingsResponse: ) -> EmbeddingsResponse:
assert self.model_store is not None assert self.model_store is not None
model = self.model_store.get_model(model_id) model = await self.model_store.get_model(model_id)
assert all(not content_has_media(content) for content in contents), ( assert all(not content_has_media(content) for content in contents), (
"Ollama does not support media for embeddings" "Ollama does not support media for embeddings"

6
llama_stack/providers/remote/inference/vllm/vllm.py
View file

@ -249,7 +249,7 @@ class VLLMInferenceAdapter(Inference, ModelsProtocolPrivate):
assert self.model_store is not None assert self.model_store is not None
if sampling_params is None: if sampling_params is None:
sampling_params = SamplingParams() sampling_params = SamplingParams()
model = self.model_store.get_model(model_id) model = await self.model_store.get_model(model_id)
request = CompletionRequest( request = CompletionRequest(
model=model.provider_resource_id, model=model.provider_resource_id,
content=content, content=content,
@ -279,7 +279,7 @@ class VLLMInferenceAdapter(Inference, ModelsProtocolPrivate):
assert self.model_store is not None assert self.model_store is not None
if sampling_params is None: if sampling_params is None:
sampling_params = SamplingParams() sampling_params = SamplingParams()
model = self.model_store.get_model(model_id) model = await self.model_store.get_model(model_id)
# This is to be consistent with OpenAI API and support vLLM <= v0.6.3 # This is to be consistent with OpenAI API and support vLLM <= v0.6.3
# References: # References:
# * https://platform.openai.com/docs/api-reference/chat/create#chat-create-tool_choice # * https://platform.openai.com/docs/api-reference/chat/create#chat-create-tool_choice
@ -397,7 +397,7 @@ class VLLMInferenceAdapter(Inference, ModelsProtocolPrivate):
task_type: Optional[EmbeddingTaskType] = None, task_type: Optional[EmbeddingTaskType] = None,
) -> EmbeddingsResponse: ) -> EmbeddingsResponse:
assert self.client is not None assert self.client is not None
model = self.model_store.get_model(model_id) model = await self.model_store.get_model(model_id)
kwargs = {} kwargs = {}
assert model.model_type == ModelType.embedding assert model.model_type == ModelType.embedding