Skip to content

[ci] add Qwen3.5 Dense/MoE models accuracy validation and benchmark tests for atom-plugined sglang#700

Open
wanzhenchn wants to merge 2 commits intoROCm:mainfrom
wanzhenchn:ci/support_qwen3.5
Open

[ci] add Qwen3.5 Dense/MoE models accuracy validation and benchmark tests for atom-plugined sglang#700
wanzhenchn wants to merge 2 commits intoROCm:mainfrom
wanzhenchn:ci/support_qwen3.5

Conversation

@wanzhenchn
Copy link
Copy Markdown
Contributor

@wanzhenchn wanzhenchn commented May 6, 2026

Motivation

  • Add ATOM-plugined SGLang accuracy coverage for the new Qwen3.5 dense(27B) and MoE(35B-A3B) models.
  • Add Qwen3.5-397B-A17B-FP8 TP4/TP8 benchmark cases on MI355

@wanzhenchn wanzhenchn force-pushed the ci/support_qwen3.5 branch 3 times, most recently from ea598c1 to 82c8443 Compare May 7, 2026 02:06
@wanzhenchn wanzhenchn force-pushed the ci/support_qwen3.5 branch 2 times, most recently from 24175af to 69f279f Compare May 8, 2026 10:05
@wanzhenchn wanzhenchn changed the title [ci] add Qwen3.5 Dense/MoE models accuracy validation for atom-plugined sglang [ci] add Qwen3.5 Dense/MoE models accuracy validation and benchmark tests for atom-plugined sglang May 8, 2026
@wanzhenchn wanzhenchn force-pushed the ci/support_qwen3.5 branch from 69f279f to 59e1bf2 Compare May 8, 2026 10:09
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant