Repository navigation
fix(vllm): deliver a base launch's generated output without an adapter selection - #67
Merged
Merged
Conversation
…r selection Signed-off-by: Feathbow <feathbow@gmail.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Why
A vLLM base launch starts the engine without LoRA. It served scoring, but generation failed at the first completed output: delivery decoded the adapter selection unconditionally, and a base launch has none. Generating with a model's base alone, for example to evaluate a distillation teacher, was therefore impossible.
What
Delivery treats a missing selection as no adapter, as scoring does, and still refuses an output that carries any adapter.
Validation
The vLLM worker suite passed on a machine with vLLM installed, including a new test of both cases; restoring the old decoding makes it fail.