feat(dflash): support dflash - #258
Open
RuixiangMa wants to merge 1 commit into
Open
Conversation
RuixiangMa
force-pushed
the
feat/dflash-support
branch
2 times, most recently
from
August 29, 2026 09:18
a731f96 to
b83e823
Compare
RuixiangMa
force-pushed
the
feat/dflash-support
branch
from
August 29, 2026 13:23
b83e823 to
9f0a136
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Implements DFlash v1 block-diffusion speculative decoding for the FreeToken inference engine. DFlash v1 uses a lightweight block-diffusion draft model to generate draft tokens in parallel, verified by the target model, achieving lossless acceleration on MoE models.
Usage
ft serve --model-path Qwen/Qwen3.6-35B-A3B
--speculative-algorithm dflash
--speculative-draft-model-path /cache/models/Qwen3.6-35B-A3B-DFlash
--speculative-dflash-block-size 8
--moe-backend fused --attention-backend fa --cache-type naive
Result
CC @jason-fxz @andy-yang-1