🤯 炸裂！Qwen 3.0-8B从BF16 → 12位完全无损压缩居然成功了！打脸压缩一定有损的言论！
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.
🤯 炸裂！Qwen 3.0-8B从BF16 → 12位完全无损压缩居然成功了！打脸压缩一定有损的言论！ @AlpinDale 基于 Together AI 的 NyoomFloat12，已中 NeurIPS 2026 实现： 🔹99.7%权重指数高4位全是0111，直接砍掉不存，解码时瞬间还原 🔹仅0.3%异常值用escape buffer精确保存，比特级无损 🔹解码后与原BF16逐比特一致，输出结果完全相同 🔹实测内存：Qwen3-8B省21%（3.2GiB），30B-A3B省24.4%（13.88GiB） 🔹吞吐提升最高+17.8%，还省带宽，速度反而更快 🔹vLLM已有PR支持在线量化，直接可用 以前觉得“压缩就一定有损”，现在被彻底打脸。 无损还能加速，这才是真正的降维打击。 #NyoomFloat12 #NF12 #Qwen #LLM #AI推理加速 #NeurIPS2026 #无损压缩 #vLLM #阿里巴巴
Posted by SuSu_酥酥👅 (27k followers) 42 min ago · 13 likes · 1.2k views · view the original post on X. Kept by the Dev Radar as AI dev tools. Tools mentioned: nf12.
More dev work like this
- Running coding agents is easy. Tracking what went wrong is harder. — @DanKornas
- The in-app browser will also be able to inherit your cookies and passwords if you opt in… — @Teknium
- Frontend code / backend code / code after the AI code review (w @karrisaarinen, @zeeg,… — @GergelyOrosz
- Tip: Switch Claude Code's output style to Concise to get short, straight-to-the-point… — @addyosmani
- 写长篇故事，别再靠一堆 Prompt 硬撑了。 — @bkdgiffug
- People still don't get how much Hermes does. — @HermesWatcher
- 100% of the code was written locally on a 16GB RTX GPU using gemma-4-12b-qat, a model… — @TeksEdge
- Missing an agent tool? Nodecode can write it and keep it. — @DanKornas
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. 31.9k posts from 5k X accounts over the last 21 days, 9.6k tools, 12 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-07 08:37 UTC. Full method.