Skip to content

Conversation

gongshaotian
Copy link
Collaborator

@gongshaotian gongshaotian commented Oct 16, 2025

Motivation

When used in CUDAGraph+MTP+EP scenarios, Draft Models on different GPUs may run empty, causing all_gather() to receive an error when obtaining step_use_cudagraph.

Modifications

Draft Model directly uses the step_use_cudagraph Flag of Target Model

Usage or Command

Accuracy Tests

Checklist

  • Add at least a tag in the PR title.
    • Tag list: [[FDConfig],[APIServer],[Engine], [Scheduler], [PD Disaggregation], [Executor], [Graph Optimization], [Speculative Decoding], [RL], [Models], [Quantization], [Loader], [OP], [KVCache], [DataProcessor], [BugFix], [Docs], [CI], [Optimization], [Feature], [Benchmark], [Others], [XPU], [HPU], [GCU], [DCU], [Iluvatar], [Metax]]
    • You can add new tags based on the PR content, but the semantics must be clear.
  • Format your code, run pre-commit before commit.
  • Add unit tests. Please write the reason in this PR if no unit tests.
  • Provide accuracy results.
  • If the current PR is submitting to the release branch, make sure the PR has been submitted to the develop branch, then cherry-pick it to the release branch with the [Cherry-Pick] PR tag.

Copy link

paddle-bot bot commented Oct 16, 2025

Thanks for your contribution!

freeliuzc
freeliuzc previously approved these changes Oct 16, 2025
@gongshaotian gongshaotian changed the title [CUDAGraph] Fix the bug of CUDAGraph + MTP + EP [Graph Optimization][Speculative Decoding] Fix the bug of CUDAGraph + MTP + EP Oct 17, 2025
freeliuzc
freeliuzc previously approved these changes Oct 17, 2025
@Jiang-Jia-Jun Jiang-Jia-Jun merged commit 528c557 into PaddlePaddle:develop Oct 20, 2025
22 of 27 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants