feat(moe): NSP-blocked expert dispatch for Qwen3MOE and GPT-OSS prefill#935
Open
feat(moe): NSP-blocked expert dispatch for Qwen3MOE and GPT-OSS prefill#935
Conversation
ochougul
reviewed
Apr 23, 2026
Add expert-blocked NSP-parallel prefill forward to QEffPrefillChunkedQwen3MoeSparseMoeBlock and QEffPrefillOnlyChunkedGptOssMLP. Controlled via EXPERT_BLOCKING_NUM_NSP env var. Fix CtxScatterFunc3D/CtxGatherFunc3D eager forward for INT32_MAX sentinel handling. Add disagg-mode tests for both models with tiny configs. Signed-off-by: vbaddi <vbaddi@qti.qualcomm.com>
Signed-off-by: vbaddi <vbaddi@qti.qualcomm.com>
Signed-off-by: vbaddi <vbaddi@qti.qualcomm.com>
Signed-off-by: vbaddi <vbaddi@qti.qualcomm.com>
Signed-off-by: vbaddi <vbaddi@qti.qualcomm.com>
…prefill Signed-off-by: vbaddi <vbaddi@qti.qualcomm.com>
vbaddi
force-pushed
the
feat/prefill_moe
branch
from
a0fe82c to
6b049bc
Compare
Signed-off-by: vbaddi <vbaddi@qti.qualcomm.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
4 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Adds NSP-parallel expert-blocked dispatch to the chunked prefill MoE path for Qwen3MOE and GPT-OSS, replacing the sequential per-expert loop with a batched packed-prefix approach.
Falls back to the original per-expert loop if
num_experts % EXPERT_BLOCKING_NUM_NSP !=0.EXPERT_BLOCKING_NUM_NSP=2 pytest tests/transformers/models/test_moe_prefill_blocked.py -vUpdate (0429):
export EXPERT_BLOCKING_PACKED_CHUNK_SIZE=256for chunk PL of 512