DeepSeek V4 Flash 0731 user benchmark reports 700pp/s, 18tg/s
Testing DeepSeek V4 Flash 0731 on a Threadripper 5965WX with 512GB RAM, a user reports 700pp/s prompt processing and 18tg/s generation using `-b 8192 -ub 8192 --cpu-moe` flags, up from 140pp/s with the default config.
1 source
Daily brief
Get tomorrow's AI brief in your inbox
More stories today
- Z.ai CEO Jie Tang: GLM 5.3 gains come from RL, not parameter count
- New tool adds 14 skills to Claude Code and Cursor for Markdown diagrams
- Tool turns Claude into a team of AI employees on your Mac
- GOP panics over Big Tech ties as Trump shifts on AI regulation
- Ethan Mollick: Claude's skill creator beats ChatGPT for reusable skills