Open
Conversation
Introduce a new speculative decoding algorithm (ASYNC_SPEC) where the draft model runs on a dedicated GPU in a separate process, communicating via multiprocessing Pipe. The key innovation is a tree cache mechanism on the draft GPU that pre-populates speculative continuations in the background. The verify step reuses SGLang's existing EagleVerifyInput infrastructure (topk=1 chain verification), so acceptance/rejection logic and KV cache management are identical to standard speculative decoding. New files: - async_spec/handshake.py: IPC protocol (SpecRequest/SpecResponse) - async_spec/verify.py: Verification algorithm (greedy + stochastic) - async_spec/tree_utils.py: Tree construction helpers - async_spec/async_draft_runner.py: Dedicated-GPU draft process with tree cache - async_spec/async_spec_worker.py: Scheduler-side orchestrator Modified files: - spec_info.py: ASYNC_SPEC enum + is_async_spec() + create_worker() - server_args.py: 6 new CLI args + validation - schedule_batch.py: recovery_token_id, last_spec_step_accepted_len on Req - scheduler.py: _init_async_spec_worker() to spawn draft process - cuda_graph_runner.py: Capture TARGET_VERIFY graphs for ASYNC_SPEC Co-Authored-By: Claude Opus 4.6 (1M context) <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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Motivation
Modifications
Accuracy Tests
Benchmarking and Profiling
Checklist
Review Process
/tag-run-ci-label,/rerun-failed-ci,/tag-and-rerun-ci