SANA-WM: Efficient Minute-Scale World Modeling with Hybrid Linear Diffusion Transformer Paper β’ 2605.15178 β’ Published 8 days ago β’ 80
Running 166 The ultimate guide to RL environments: building and scaling them in the LLM era π 166 Building and scaling RL environments for LLM training
view article Article ML Intern Takes Our Post-Training Internship Test cmpatino β’ 29 days ago β’ 31
Running RL Featured 79 Survival Island Game π 79 Watch an AI agent survive on a procedurally generated island
Running Featured 84 Distilling 100B+ Models 40x Faster with TRL π 84 TRL distillation for 100B+ teachers, 40x faster
Gemma 4 Collection Gemma 4 is Google's new model family including including E2B, E4B, 26B-A4B, and 31B. β’ 28 items β’ Updated 30 days ago β’ 190
Transformers.js V4 demos Collection A collection of demos built with Transformers.js V4 β’ 24 items β’ Updated Apr 16 β’ 58
Running on CPU Upgrade 236 The Synthetic Data Playbook: Generating Trillions of the Finest Tokens π 236 Explore synthetic data experiments on a virtual bookshelf
Jackrong/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled Image-Text-to-Text β’ 28B β’ Updated Apr 6 β’ 199k β’ β’ 2.85k