Right when I finished downloading the full weights of DeepSeek V4.1 flash for running it…
This is a dev post classified by AGK Intelligence as AI dev tools (a tool drop), kept by the Dev Radar because it carries real work, not commentary.
Right when I finished downloading the full weights of DeepSeek V4.1 flash for running it on four Sparks, @NVIDIAAI released an NVFP4 quant that is 81GB smaller in size, basically with no quality loss. Downloading that one now 👇
Posted by Javier • priv/acc (4.9k followers) 1 h ago · 2 likes · 133 views · view the original post on X. Kept by the Dev Radar as AI dev tools.
More dev work like this
- The best world a local AI agent built on each memory class, from 512 GB down to 16 GB.… — @niklaslenz_ai
- Opus 5.5 built a real 3D skeleton that draws itself down to pixels! We then take a 2d… — @rehan_shei
- Day 31/ — @theo
- AI 圈最狠的“脑外科手术”工具来了 🧠🔪 — @mylifcc
- A diagram drawn in JavaScript using @poteto's /statechart skill. The life of one animal… — @kevin_t_ngo
- Gemini 4 Carbon scores 53 FCI and 54 pass rate on FrontierCS 2. It’s an impressive jump… — @wenhaocha1
- Clover’s speed caught my attention. Its task results deserve attention too. — @TeksEdge
- 🤔我靠！微软悄咪咪开源4B AI代理Mage！66%逼近人类72.4%，碾压6倍大模型，本地可跑隐私拉满！ — @NFT_Chen
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 31.1k posts from 5k X accounts over the last 21 days, 8.8k tools, 12 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-10-31 09:46 UTC. Full method.