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

Fix Meta reference GPU implementation (#663)

Browse source 
By performing in-place mutations, we lost. Never in life do that.
This commit is contained in:
Ashwin Bharambe 2024-12-19 14:09:45 -08:00 • committed by GitHub
parent f19eb8eee3
commit 540fc4d717
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
2 changed files with 15 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

9
llama_stack/providers/utils/inference/prompt_adapter.py
View file

@ -94,9 +94,14 @@ async def convert_request_to_raw(
d = m.model_dump()
d["content"] = content
messages.append(RawMessage(**d))
request.messages = messages
d = request.model_dump()
d["messages"] = messages
request = ChatCompletionRequestWithRawContent(**d)
else:
request.content = await interleaved_content_convert_to_raw(request.content)
d = request.model_dump()
d["content"] = await interleaved_content_convert_to_raw(request.content)
request = CompletionRequestWithRawContent(**d)
return request