-
Notifications
You must be signed in to change notification settings - Fork 171
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Livepool llm rebase #3178
Livepool llm rebase #3178
Conversation
dac792f
to
2095440
Compare
2095440
to
a3d7462
Compare
@@ -23,6 +23,7 @@ type AI interface { | |||
Upscale(context.Context, worker.GenUpscaleMultipartRequestBody) (*worker.ImageResponse, error) | |||
AudioToText(context.Context, worker.GenAudioToTextMultipartRequestBody) (*worker.TextResponse, error) | |||
SegmentAnything2(context.Context, worker.GenSegmentAnything2MultipartRequestBody) (*worker.MasksResponse, error) | |||
LlmGenerate(context.Context, worker.GenLlmFormdataRequestBody) (interface{}, error) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@kyriediculous, @ad-astra-video can we rename this to LLMGenerate?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes sounds good
line := scanner.Text() | ||
if strings.HasPrefix(line, "data: ") { | ||
data := strings.TrimPrefix(line, "data: ") | ||
if data == "[DONE]" { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@kyriediculous why do we need both status DONE
and a done boolean?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Per discord message, SSE streams end with a [DONE]
message. This comment can be addressed in update to Livepool LLM PR.
Closing, Livepool LLM PR will be used to take this across the finish line. This PR was created to hopefully save some time on the updates needed to get there since I had to make these changes to test the pipeline end-to-end. |
What does this pull request do? Explain your changes. (required)
Rebases Livepool LLM PR
Specific updates (required)
How did you test each of these updates (required)
stream=false
andstream=true
curl -N -X POST http://127.0.0.1:6666/llm-generate -F "prompt=tell me a story about Mario taking a beach vacation" -F "model_id=meta-llama/Meta-Llama-3.1-8B-Instruct" -F max_tokens=256 -F stream=true
Does this pull request close any open issues?
Checklist:
make
runs successfully./test.sh
pass