Running on CPU Upgrade 757 The Ultimate MoE Training Handbook 🧩 757 A hands-on guide to training Mixture-of-Experts models
Nemotron Labs Dory Collection Open looped language models with elastic depth, serving different compute budgets from a single checkpoint. • 4 items • Updated 7 days ago • 30
0bserverx/Qwen3.8-27B-Heretic-Abliterated-Uncensored-GGUF Text Generation • 27B • Updated Sep 20 • 1.2M • 910
yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF Text Generation • 12B • Updated Jun 19 • 76.3k • 1.79k
view post Post 4,743 TRL now supports MOPD (Multi-Teacher On-Policy Distillation) natively❗️Many frontier open models use it. labs train one RL expert per domain and then distill them all into one modelBlog with what it is, who uses it and an experiment you can run on HF Jobs with one commandhttps://huggingface.co/blog/sergiopaniego/mopd-trl See translation 🔥 8 8 👍 4 4 + Reply