[Feature request] decode attention support mtp-3
Open
Nobody has claimed this yet.
- Dominant language
- C++
- Stars
- 1.2k
- Forks
- 142
- Avg merge
- 23h 15m
- Merged PRs (30d)
- 8
Description
now hpc-ops support mtp-1 and mtp-2, but sglang usually recommend using --speculative-num-steps 3 for better throughput
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by locating the existing MTP-1 and MTP-2 decode-attention support and the entry point handling speculative steps. Determine how MTP-3 should fit that path, then verify that configuring three speculative steps works and preserves the intended throughput behavior.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp
- Domain
- ai-infra-agents, performance
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 45/100