hanxiao / DeepSeek-V4.1-Flash-1017-FP8-abliterated-4x-RTX-PRO-6000
git clone https://github.com/hanxiao/DeepSeek-V4.1-Flash-1017-FP8-abliterated-4x-RTX-PRO-6000.git
Abliterated DeepSeek-V4.1-Flash-1017 (FP8 dense, FP4 experts) on 4x RTX PRO 6000 with vLLM-Moet r21 and DSpark: 371-376 tok/s single-stream agent decode at 30k-120k context, 1M context, Flex-start GCE VM
—
stars, 24h
+25
stars, 7 days · #8,513
+25
stars, 30 days · #24,724
25
★ total · +25% this week
4
forks · 0 open issues
Star history starts 2026-10-27; the chart appears after the second daily snapshot.
LanguageShell
License—
Created2026-10-23 (4d ago)
Last pushtoday
OwnerUser
GitHub topics
abliteratedblackwelldeepseekdeepseek-v4flex-startfp4fp8gcpllm-inferencertx-pro-6000speculative-decodingvision-language-modelvllmvllm-moet
Links
- GitHub: github.com/hanxiao/DeepSeek-V4.1-Flash-1017-FP8-abliterated-4x-RTX-PRO-6000
- Releases · Issues · Hacker News mentions
README
Loading README…