Event ·Multi-source

Tri Dao 团队开源内核优化智能体，沉淀两年推理引擎调优经验

FlashAttention 作者 Tri Dao 宣布开源一个面向内核与推理引擎优化的智能体，将其团队两年积累的 GPU kernel 调优经验封装成可自动运行的 agent 工作流，目前已在其生产环境中使用。项目定位于自动生成和优化 kernel 代码，替代部分人工调优流程。

Current read

FlashAttention 作者 Tri Dao 把团队两年的 GPU kernel 调优经验做成了自动优化智能体，写高性能算子的可以试试。

4 sources57° AI heatLast updated 2026/11/5 19:01:00

Evidence

4 sources

Primary sourceTri Dao (FlashAttention)

2026/11/5 19:01:00View source

Related source 1Together AI

2026/11/6 22:56:00View source

Related source 2arXiv: DeepSeek

2026/11/5 05:58:36View source

Related source 3marktechpost

2026/11/6 17:28:04View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.

Open the story