fix: pass weight shape instead of tensor to should_use_deepgemm_for_fp8_linear - #2574
Merged
Conversation
…fp8_linear` vLLM 0.20.0 changed `should_use_deepgemm_for_fp8_linear` to expect a `tuple[int, int]` shape rather than the weight tensor. Passing the FP8 tensor caused `weight_shape[0] % N_MULTIPLE` to fail with `NotImplementedError: "remainder_cuda" not implemented for 'Float8_e4m3fn'` during weight loading. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> Signed-off-by: Kajal Jain <kajalj@nvidia.com>
|
Auto-sync is disabled for ready for review pull requests in this repository. Workflows must be run manually. Contributors can view more details about this message here. |
Contributor
Author
|
/ok to test 77eafab |
Contributor
|
The relevant vLLM API is changed in vllm-project/vllm@2e9034c#diff-9a1a1730a1fc1868db95c85ee80170d8ab45f4db59d372a81225c6a6268cfc27 so this is reasonable fix. LGTM. |
guyueh1
approved these changes
May 27, 2026
vigneshwaran
pushed a commit
to vigneshwaran/RL
that referenced
this pull request
May 28, 2026
…p8_linear (NVIDIA-NeMo#2574) Signed-off-by: Kajal Jain <kajalj@nvidia.com> Co-authored-by: Claude Opus 4.6 <noreply@anthropic.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
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
should_use_deepgemm_for_fp8_linearin vLLM 0.20.0 expectsweight_shape: tuple[int, int], but NeMo-RL was passinglayer.weight(the full FP8 tensor)NotImplementedError: "remainder_cuda" not implemented for 'Float8_e4m3fn'during weight loading when DeepGEMM is enabledlayer.weight.shapeinstead oflayer.weightRoot cause
The vLLM 0.17→0.20 upgrade (#2384) changed the
should_use_deepgemm_for_fp8_linearAPI from accepting a tensor to accepting a shape tuple. The call site inmaybe_post_process_fp8_weight_blockwas not updated.🤖 Generated with Claude Code