精选理由
Blackwell 上推理快了 31%
Together AI 发布了基于 Blackwell 的推理引擎，在 AgentPerf 基准测试中，其 TPS 比次快的开源引擎高出 31%。该引擎通过为 Blackwell 的 Tensor Core 指令定制内核实现性能提升。Cursor 已将其实时编程助手部署在该推理栈上。Together AI 在推文中详细介绍了构建过程。
原文 · Together AI
The case for Blackwell in production agent infrast…
The case for Blackwell in production agent infrastructure just got cleaner.
@ArtificialAnlys AgentPerf gives the hardware picture.
Together's coding agent benchmarks give the inference picture: 31% more TPS than the next-fastest OSS engine on the same hardware, through custom kernels built for Blackwell's Tensor Core instructions.
Cursor runs their real-time coding agents on this stack. Learn more about how we built it in the 🧵