phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-12-17 02:22:38 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

make model registery a helper for ollama and vllm

Browse source
This commit is contained in:
Dinesh Yeduguru 2024-11-13 12:24:18 -08:00
parent 3b68e6cbbe
commit f63d51963d
2 changed files with 13 additions and 10 deletions
Download patch file Download diff file Expand all files Collapse all files

10
llama_stack/providers/remote/inference/ollama/ollama.py
View file

@ -71,9 +71,9 @@ model_aliases = [
]
class OllamaInferenceAdapter(Inference, ModelRegistryHelper, ModelsProtocolPrivate):
class OllamaInferenceAdapter(Inference, ModelsProtocolPrivate):
def __init__(self, url: str) -> None:
ModelRegistryHelper.__init__(
self.model_register_helper = ModelRegistryHelper(
self,
model_aliases=model_aliases,
)
@ -203,7 +203,9 @@ class OllamaInferenceAdapter(Inference, ModelRegistryHelper, ModelsProtocolPriva
else:
input_dict["raw"] = True
input_dict["prompt"] = chat_completion_request_to_prompt(
request, self.get_llama_model(request.model), self.formatter
request,
self.model_register_helper.get_llama_model(request.model),
self.formatter,
)
else:
assert (
@ -283,7 +285,7 @@ class OllamaInferenceAdapter(Inference, ModelRegistryHelper, ModelsProtocolPriva
raise NotImplementedError()
async def register_model(self, model: Model) -> Model:
model = await super().register_model(model)
model = await self.model_register_helper.register_model(model)
models = await self.client.ps()
available_models = [m["model"] for m in models["models"]]
if model.provider_resource_id not in available_models: