DeepSeek V4-Flash (284B MoE) at 33 tok/s single / 68 tok/s aggregate on 2× RTX 3090 + a used quad-Xeon DDR4 server — full config

Ran DeepSeek V4-Flash-0731 — the full official checkpoint, not a re-quant — on commodity used hardware. Sharing because I couldn't find anyone else publishing Ampere results for this engine. Edit / update: a commenter called out that hybrid CPU-GPU posts always publish decode and never prefill

中文全文翻译正在补齐，可先查看下方英文原文。

Ran DeepSeek V4-Flash-0731 — the full official checkpoint, not a re-quant — on commodity used hardware. Sharing because I couldn't find anyone else publishing Ampere results for this engine. Edit / update: a commenter called out that hybrid CPU-GPU posts always publish decode and never prefill