feat: upgrade vllm amzn2023 to deepep v2 over efa - #6730
Draft
jinyan-li1 wants to merge 16 commits into
Draft
Conversation
jinyan-li1
force-pushed
the
deepep-v2/vllm-amzn2023
branch
from
September 11, 2026 18:41
059bf85 to
13b4dae
Compare
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Upgrades the vLLM AL2023 EC2 + SageMaker images to DeepEP v2 over EFA.
Changes
0.27.1→0.29.0(NCCL 2.31.2 segfault fix)amazon-contributing/DeepEP@874779c, built as a wheel from the fork (drops the upstreaminstall_python_libraries.shpath)2.30.7→2.31.2(pip wheel; required for the GIN/EFA backend)nvidia-nvshmem-cu13==3.3.242.5.2userspace libs built from source (net-new for vLLM)1.47.0→1.50.0(auto--disable-ngc)instanttensor+modelexpressdlc_minor_version4 → 5Validation: DeepEP internode (
test_ep.pyon 2x p5en) pending capacity; gated test wiring to follow. Draft until validated.