Qwen3.6-35B-A3B-heretic NVFP4 + DFlash speculative decoding on DGX Spark (GB10/sm_121a). Source-built vLLM image + 7 patches + comprehensive deployment guide.

Fully automatic censorship removal for language models

🧠「大模型」2小时完全从0训练64M的小参数LLM！Train a 64M-parameter LLM from scratch in just 2h!

A monitoring hub that watches popular open-source repositories and uses AI to detect when commits are patching security vulnerabilities - often before a CVE is even assigned. Findings are published…

[ICML 2026] Reproduction package for AutoBaxBuilder, a framework to build new BaxBench tasks from scratch

Model Context Protocol Servers