Thomas Wolf· @Thom_Wolf · X·· 4 小时前AI 评分69
AI 导读
Thomas Wolf 转评 DeepSeek 宣布将其生产推理框架 DeepInfer 开源，该框架支撑 DeepSeek-V4.1-Flash 的高吞吐低延迟推理，代码以 Apache 2.0 协议发布在 http://github.com/deepseek-ai/DeepInfer。他同时列举 DeepSeek 此前开源的 FlashMLA（H800 上 580 TFLOPS attention）、DeepEP（MoE 通信）、DeepGEMM（1350+ FP8 TFLOPS 矩阵乘）和 3FS 文件系统，称其开源势头令人敬佩。
正文 · 原文
DeepSeek's open-source streak is getting absurd:
- FlashMLA: 580 TFLOPS attention on H800
- DeepEP: MoE comms
- DeepGEMM: 1350+ FP8 TFLOPS matmul
- 3FS: they built an entire file system
- and today: the full inference engine behind V4.1-Flash, Apache 2.0
Mad respect
🚀 Introducing DeepInfer, our in-house inference framework, now fully open source! Powering DeepSeek-V4.1-Flash at scale with high throughput and low latency. We’re sharing our production stack to help the community lower compute costs, reduce energy use, and make AI more affordable. 🔗 http://github.com/deepseek-ai/DeepInfer 1/4在 X 查看被引用的帖子
来源：Thomas Wolf · x.com