Skip to content
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

[Bugfix] Fix GGUF inference with FP16 unquantized checkpoint #10675

Merged
merged 5 commits into from
Nov 27, 2024

Conversation

Isotr0py
Copy link
Collaborator

@Isotr0py Isotr0py commented Nov 26, 2024

FIX #10600 (link existing issues this PR will resolve)

Signed-off-by: Isotr0py <2037008807@qq.com>
Signed-off-by: Isotr0py <2037008807@qq.com>
Signed-off-by: Isotr0py <2037008807@qq.com>
Signed-off-by: Isotr0py <2037008807@qq.com>
Copy link

👋 Hi! Thank you for contributing to the vLLM project.
Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can do one of these:

  • Add ready label to the PR
  • Enable auto-merge.

🚀

Copy link
Member

@mgoin mgoin left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM - should there be asserts/checks on the qweight_type in ggml_mul_mat_vec_a8 and ggml_mul_mat_a8?

Signed-off-by: Isotr0py <2037008807@qq.com>
Copy link
Collaborator Author

@Isotr0py Isotr0py left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

should there be asserts/checks on the qweight_type in ggml_mul_mat_vec_a8 and ggml_mul_mat_a8?

Sure, I have added qweight_type checking for ggml_mul_mat_vec_a8 and ggml_mul_mat_a8 with a fallback. PTAL!

Copy link
Member

@mgoin mgoin left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Excellent, I love the explicit sets

@mgoin mgoin added the ready ONLY add when PR is ready to merge/full CI is needed label Nov 27, 2024
@simon-mo simon-mo merged commit b98c62b into vllm-project:main Nov 27, 2024
58 of 60 checks passed
@Isotr0py Isotr0py deleted the fix-gguf-fp16 branch November 28, 2024 02:28
afeldman-nm pushed a commit to neuralmagic/vllm that referenced this pull request Dec 2, 2024
…oject#10675)

Signed-off-by: Isotr0py <2037008807@qq.com>
Signed-off-by: Andrew Feldman <afeldman@neuralmagic.com>
sleepwalker2017 pushed a commit to sleepwalker2017/vllm that referenced this pull request Dec 13, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
ready ONLY add when PR is ready to merge/full CI is needed
Projects
None yet
Development

Successfully merging this pull request may close these issues.

[Bug]: GGUF Model Output Repeats Nonsensically
3 participants