vLLM 团队庆祝 DeepSeek-V4.1-Flash-1017 发布,达成 1.3B tokens/美元 TCO
vLLM 官方晒出 DeepSeek-V4.1-Flash-1017 的服务成绩:每美元 1.3B tokens,单卡 830K tok/s,跑智能体循环很省钱。
vLLM 项目团队祝贺 DeepSeek 发布 DeepSeek-V4.1-Flash-1017,称其分数更高且输出更短,适合长智能体循环场景。在 SemiAnalysis AgentX 基准上,vLLM 服务该模型达到 1.3B tokens/美元的 TCO,单 GPU 吞吐 830K tok/s(含缓存、输入和输出),同时保证每用户 100 tok/s。团队还表示自第一天起已实现 5 倍智能体吞吐提升。
🎉 Congrats @deepseek_ai on DeepSeek-V4.1-Flash-1017! Better scores and shorter outputs make it a great fit for long agent loops. Making each step fast is the serving side's job.
On DeepSeek-V4.1-Flash, vLLM delivers 1.3B tokens per dollar TCO on SemiAnalysis AgentX: 830K tok/s per GPU (cached + input + output) at 100 tok/s per user. 🚀
It's been a fun few weeks building with the community. Chapter one, 5x agentic throughput since day 0: 🔗 https://t.co/2ZE4ZPrsBx
Next chapter coming soon. 👀