Uncensored deepseek V4 pro distill that runs on your laptop
Heretic stripped the filters off a deepseek v4 pro distill of qwen3.5 9b. free weights that fit a normal pc how to get it: 1. the uncensored model 2. ready-to-run gguf files
حد عمل ريبو على GitHub بيخليك تستخدم Claude Code مجاني وللأبد 🤯
بيحوّل طلبات Claude Code لأكتر من 50 provider، كتير منهم مجاني، زي DeepSeek وKimi وGemini وNVIDIA NIM.
Stormlit Ocean Lighthouse: 🏆 DeepSeek V4.1 Flash max: 7.99 · $0.14 · 18m36s
Gemini 4 Carbon max: 6.65 · $1.15 · 21m40s
VLLM merged a hand-written decode path for GLM-5.3-Flash on 4x GB300
Median inter-token latency drops from 5.15 ms to 2.93 ms at batch 1. At batch 64 the same change buys 13.92 to 12.00 ms. That curve is the story. How it works: CuTe DSL kernels for every sublayer, fused
You have reached the end of the archive
All of deepseek