mudler / mudler/vllm.cpp

GFX1100-TG200: small-N GEMV geometry levers (VT_GEMV_WARPS, VT_GEMV_SPLITK) — closed negative

Open
#2,887 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
C++
Stars
423
Forks
53
Avg merge
20h 26m
Merged PRs (30d)
310

Description

Row: GFX1100-TG200

Summary

Small-N GEMV geometry levers: sweep warp count (VT_GEMV_WARPS) and split-K (VT_GEMV_SPLITK) for decode GEMV shapes. Closed negative in the original campaign — warps/split-K win warm benches but wash in the engine.

Benchmark

A/B interleaved, 5 pairs, Qwen3.5-4B Q4_K_M, 256 tokens, temp 0, seed 0:

A (T25 chain) B (T25+T37) Delta
Median 32.40 32.38 -0.09%

Noise — confirms the original campaign's negative disposition.

Token identity

PASS — identical output to T25 chain.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Review the small-N GEMV benchmark campaign for the VT_GEMV_WARPS and VT_GEMV_SPLITK levers on GFX1100-TG200. Compare warm-benchmark results with engine-level measurements and verify token identity; the work is done only if a geometry setting produces a repeatable engine improvement, since the reported campaign was negative.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp
Domain
performance
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Active
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.