Open-weight & owned stack · Open lab releases

DeepSeek V4.1 Flash Temporary Build Shows High Speed

Sep 9, 2026

A temporary DeepSeek V4.1 Flash test build reportedly processed up to 400 tokens per second on 3D coding tasks.

Read the original at geeky-gadgets.comOpens the publisher's site in a new tab

More in Open-weight & owned stack

Z.ai publishes GLM-5.3 open weightsSep 8 MiniMax H3 highlighted as open-weight multimodal modelSep 7 InternLM releases 397B multimodal Intern-S2 modelSep 13 DeepSeek Model Nearly Matches GPT-6 Astra DesignSep 10 DeepSeek presents V4 FlashSep 8