phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-08-06 10:42:39 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

fix again vllm

Browse source
This commit is contained in:
Xi Yan 2025-01-17 17:46:42 -08:00
parent 97053ba9d6
commit f2ddf02fb6
2 changed files with 3 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

3
llama_stack/providers/remote/inference/vllm/vllm.py
View file

@ -177,7 +177,8 @@ class VLLMInferenceAdapter(Inference, ModelsProtocolPrivate):
if isinstance(request, ChatCompletionRequest): if isinstance(request, ChatCompletionRequest):
if media_present: if media_present:
input_dict["messages"] = [ input_dict["messages"] = [
await convert_message_to_openai_dict(m) for m in request.messages await convert_message_to_openai_dict(m, download=True)
for m in request.messages
] ]
else: else:
input_dict["prompt"] = await chat_completion_request_to_prompt( input_dict["prompt"] = await chat_completion_request_to_prompt(

2
llama_stack/providers/utils/inference/prompt_adapter.py
View file

@ -188,7 +188,7 @@ async def localize_image_content(media: ImageContentItem) -> Tuple[bytes, str]:
async def convert_image_content_to_url( async def convert_image_content_to_url(
media: ImageContentItem, download: bool = False, include_format: bool = True media: ImageContentItem, download: bool = False, include_format: bool = True
) -> str: ) -> str:
if media.url and not download: if media.url and (not download or media.url.uri.startswith("data")):
return media.url.uri return media.url.uri
content, format = await localize_image_content(media) content, format = await localize_image_content(media)