菜单
Contribute release-by-release performance summaries to vllm-omni-cookbook: per-model improvement ledgers (index.md) and cross-model release tables (SUMMARY.md). Use when documenting what improved at vX.Y.Z, updating perf deltas, or adding a new model ledger.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Add a new model to vllm-omni-cookbook with a release-by-release improvement ledger (index.md). Use when onboarding a model not yet tracked under omni/ or diffusion/.
Append improvement summary for an even vLLM-Omni release (v0.22.0, …): add vX.Y.Z to each model index.md and update SUMMARY.md. Skip odd minors. Use when v0.22.0 ships or backfilling even-release summaries.
Write optional Zhihu draft for vllm-omni-cookbook model folders. Use after retro tables exist in index.md, for Chinese republishing of performance ledgers.
| name | vllm-omni-cookbook |
| description | Contribute release-by-release performance summaries to vllm-omni-cookbook: per-model improvement ledgers (index.md) and cross-model release tables (SUMMARY.md). Use when documenting what improved at vX.Y.Z, updating perf deltas, or adding a new model ledger. |
This repo summarizes model performance improvements at every even stable release — not deployment how-tos. vLLM-Omni owns recipes and benchmarks; here we publish what changed, by how much, and why (PR links).
Even releases only: add cookbook sections for v0.18.0, v0.20.0, v0.22.0, … Skip odd minors. Deltas compare to the previous even release.
Each {category}/{model}/index.md is the model's full improvement timeline. SUMMARY.md is the per-release cross-model snapshot.
vllm-omni/benchmark_results/.vllm + vllm_omni versions, GPU SKU/count, serve args, workload dimensions.| Goal | Skill |
|---|---|
| First-time model in the cookbook | cookbook-add-model |
| New even stable release (v0.22.0, …) | cookbook-add-release |
| Optional Zhihu draft | cookbook-write-narrative |
| Category | Folder | Primary metrics |
|---|---|---|
| Omni | omni/ | TTFT, TTFP, TPOT, RTF, E2EL, throughput |
| Diffusion | diffusion/ | E2E latency (latency_mean), optional stage profiler |
vllm-omni-cookbook/
├── SUMMARY.md # per-release cross-model snapshot
├── omni/{model}/index.md
└── diffusion/{model}/
├── index.md # slim reference ledger
└── *-performance-zhihu.md # optional Chinese draft
Canonical examples: diffusion/wan2.2/, omni/qwen3-omni/.
SUMMARY.md updated if adding a release sectionindex.md and Zhihu draftSee references/templates.md for index.md, SUMMARY.md, and PR body templates.