benchmark-kernel
Guide for benchmarking FlashInfer kernels with CUPTI timing
浏览 flashinfer-ai 维护的已发布 AI 技能,比较用途与说明,并通过来源仓库了解安装方式。
已收录技能
7收录技能的仓库 Stars 合计
15,370仓库 Forks
2,197平均每个技能的仓库 Stars
2,196同一仓库中的多个技能可能共享 Stars,合计包含重复仓库数据。
Guide for benchmarking FlashInfer kernels with CUPTI timing
Tutorial for debugging CUDA crashes using API logging
Step-by-step tutorial for adding new CUDA kernels to FlashInfer
Track popular/new open-source LLMs and update docs/model_coverage.mdx with their kernel support status. Use when discovering new models to add to the coverage tracker, checking if a specific model is covered, o...
Clone SGLang, FlashInfer repositories from GitHub to tmp/. Use when setting up the project, preparing for kernel extraction, or when the user needs the source repositories.
Add pytest tests to validate reference implementations in flashinfer_trace against FlashInfer or SGLang ground truth. Use when validating kernel definitions, adding tests for new op_types, or verifying referenc...
Extract kernel schemas and definitions from SGLang model implementations with deduplication. Use when adding a new model, extracting GPU kernels (MLA, MoE, GQA, RMSNorm, GEMM), or generating Definition JSON fil...
默认按已收录数据中的 GitHub 仓库 Stars 排序。Stars 属于来源仓库,同一仓库的多个技能可能共享相同数值;它反映仓库关注度,不代表单个技能的安装量或质量评分。