Skip to content

Pull requests: vllm-project/perf-eval

New pull request New
25 Open 52 Closed
25 Open 52 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Add GB300 Slurm PD serving infrastructure and Kimi K3 PD workloads
#74 opened Sep 2, 2026 by khluu Member Loading…
Remove Kimi K2.5 workloads
#73 opened Sep 2, 2026 by khluu Member Loading…
[RFC] Capture reproducible experiment provenance
#66 opened Aug 18, 2026 by jamesETsmith Contributor • Draft
add more tests
#64 opened Aug 17, 2026 by ZJY0516 Member Loading…
2
Add Kimi K3 standalone B300 perf workload
#61 opened Aug 11, 2026 by khluu Member • Draft
Harden pipeline dependency downloads
#54 opened Jul 29, 2026 by khluu Member • Draft
Add GLM-5.2 standalone B300 SGLang benchmark
#51 opened Jul 28, 2026 by simon-mo • Draft
Add GLM-5.2 standalone B300 vLLM benchmark
#50 opened Jul 28, 2026 by simon-mo • Draft
1
Add Qwen3.6-35B-A3B NVFP4 gsm8k accuracy coverage (1xB200)
#49 opened Jul 27, 2026 by stecasta • Draft
1 task
Add Devstral-Small-2-24B gsm8k accuracy coverage (1xB200)
#48 opened Jul 27, 2026 by stecasta Loading…
1
Organize workloads into model directories
#47 opened Jul 22, 2026 by khluu Member • Draft
1
Add DeepSeek V4 Pro B200 workload
#45 opened Jul 22, 2026 by khluu Member Loading…
8
Add Kimi K2.5 B200 workload
#43 opened Jul 22, 2026 by khluu Member Loading…
4
Add workload lint + CONTRIBUTING for pre-merge validation
#41 opened Jul 21, 2026 by stecasta • Draft
Add Kimi-K2-Thinking NVFP4/INT4 gsm8k accuracy coverage (4xB200)
#40 opened Jul 21, 2026 by stecasta Loading…
Add Gemma-4-31B-IT NVFP4 gsm8k accuracy coverage (1xB200)
#39 opened Jul 21, 2026 by stecasta Loading…
1
Add xpu case
#38 opened Jul 20, 2026 by wendyliu235 Loading…
3
Switch vllm_bench to standalone vllm-bench Rust binary
#32 opened Jul 1, 2026 by khluu Member Loading…
2 of 3 tasks
Add GB300 Slurm Pyxis smoke path
#30 opened Jun 27, 2026 by khluu Member • Draft
attention backend sweep
#29 opened Jun 23, 2026 by stacyroberts • Draft
deepseek_v4_flash-b200: align serve args with recipe + fix cu13 TileLang JIT
#27 opened Jun 18, 2026 by khluu Member Loading…
Reap stale perf-eval containers before starting vLLM
#26 opened Jun 15, 2026 by khluu Member Loading…
Add MiniMax-M3 Nightly Perf & Accuracy Regression Testing on MI355X with EAGLE3 spec decode
#25 opened Jun 14, 2026 by functionstackx Loading…
2 of 3 tasks
1
NVIDIA-Nemotron-3-Ultra-550B-A55B
#18 opened Jun 11, 2026 by NickLucche Member Loading…
Add B200 workload variants for all H200-only models
#12 opened May 28, 2026 by khluu Member Loading…
1 of 2 tasks
ProTip! Find all pull requests that aren't related to any open issues with -linked:issue.