AI 导读
DeepSeek 宣布将自研推理框架 DeepInfer 完全开源，代码位于 github.com/deepseek-ai/DeepInfer。该框架在生产环境支撑 DeepSeek-V4.1-Flash 的大规模推理，主打高吞吐和低延迟，官方称希望帮助社区降低计算成本与能耗。
推荐理由
DeepSeek 公开自研推理框架的生产栈，想降低推理算力成本的团队可以参考其吞吐与延迟实践。
正文
🚀 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
来源：DeepSeek · x.com