产品46

Sciforium 在 4 块 AMD MI355X 上跑 DeepSeek-V4.1-Flash 达 1129 tokens/s

原标题:Sciforium shows DeepSeek-V4.1-Flash reaching 1,129 output tokens/sec at P90 on 4 AMD Instinct MI355X GPUs, keeping DeepSeek's original we...

Rohan Paul@rohanpaul_ai发布时间 (北京时间)

Sciforium 展示 DeepSeek-V4.1-Flash 在 4 块 AMD Instinct MI355X GPU 上 P90 输出达 1,129 tokens/sec,且沿用 DeepSeek 原始权重。

中文译文尚未完成,先为你展示原文。

Sciforium shows DeepSeek-V4.1-Flash reaching 1,129 output tokens/sec at P90 on 4 AMD Instinct MI355X GPUs, keeping DeepSeek's original weights.

The gain comes from speculative decoding done better. Sciforium's Traffic-tuned Drafter is tuned on customer traffic, so more of its proposed tokens get accepted and draft length could be raised to seven tokens.

Separately, kernel work fused small ops like mHC and projections and fused all-reduce with compute.

At that rate, 10,000 tokens is about nine seconds of generation.

not long ago, people questioned whether AMD GPUs could keep up. today, we’re setting records on them. 1,129 output tps at P90. Real production traffic, not a toy project. a little faith, a lot of engineering. here’s how we did it. https://x.com/i/article/2122759644397674496
查看原帖
DeepSeek推理能力部署工程产品更新#产品更新#推理#部署/工程#DeepSeek