phoenix-oss/llama-stack
1
0
Fork 0
forked from phoenix-oss/llama-stack-mirror
Code Pull requests Releases Packages 2 Activity Actions

Make embedding generation go through inference (#606)

Browse source 
This PR does the following:
1) adds the ability to generate embeddings in all supported inference
providers.
2) Moves all the memory providers to use the inference API and improved
the memory tests to setup the inference stack correctly and use the
embedding models

This is a merge from #589 and #598
This commit is contained in:
Dinesh Yeduguru 2024-12-12 11:47:50 -08:00 • committed by GitHub
parent a14785af46
commit 96e158eaac
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
37 changed files with 677 additions and 156 deletions
Download patch file Download diff file Expand all files Collapse all files

12
llama_stack/providers/remote/inference/together/together.py
View file

@ -31,6 +31,7 @@ from llama_stack.providers.utils.inference.openai_compat import (
from llama_stack.providers.utils.inference.prompt_adapter import (
chat_completion_request_to_prompt,
completion_request_to_prompt,
content_has_media,
convert_message_to_dict,
request_has_media,
)
@ -253,4 +254,13 @@ class TogetherInferenceAdapter(
model_id: str,
contents: List[InterleavedTextMedia],
) -> EmbeddingsResponse:
raise NotImplementedError()
model = await self.model_store.get_model(model_id)
assert all(
not content_has_media(content) for content in contents
), "Together does not support media for embeddings"
r = self._get_client().embeddings.create(
model=model.provider_resource_id,
input=[interleaved_text_media_as_str(content) for content in contents],
)
embeddings = [item.embedding for item in r.data]
return EmbeddingsResponse(embeddings=embeddings)