DeepSeek Flash 跑到约 1000 tokens 每秒，靠的是模型驱动的推理基础设施，做实时应用的人可以关注下。
DeepSeek Flash 的推理速度被实测约为每秒 1000 tokens。这一速度由模型驱动的底层基础设施优化支撑，而不是单靠硬件堆叠。对交互式应用来说，千 tokens 级别的生成速度能明显降低等待感。
~1000 TPS on deepseek flash love us some model driven model infra
同样 100 美元，不同模型差出 60 倍任务量，选模型前看这组性价比数据能省不少钱。
The benchmark gap over DeepSeek-V4-Flash-Base in the same evaluation setup makes this one of the stronger open base weights released for coding workloads, and day-0 vLLM support means teams can deploy it without custom serving work. Practitioners running RL research or domain adaptation get open base weights plus serving recipes for expert parallelism, KV offload and prefill/decode disaggregation on GB200 and H100 hardware.
Response-ending behaviour shapes how conversational and proactive an assistant feels, and this benchmark turns that personality trait into a measurable signal. Product teams choosing a model for customer-facing chatbots can use these figures to match tone, since a 34% question rate reads very differently from a 0.1% one. The spread between MiMo-V2.6-Pro and DeepSeek V4.1 Flash also shows two Chinese labs optimising for opposite interaction styles.
DeepSeek Harness 日活破 154 万了，插件已有 3.5 万个，官方还打算做插件市场给作者发 Token 激励，做插件的朋友可以关注。
装了 Hermes 的可以抄作业：8 个实测插件，从中文包、微信读书同步到按模型算 token 花费，都是能直接用的。
