Muse Spark Open Collection Muse Spark Open model. A 816B parameter model with 42B Active parameters. BF16 and NVFP4 Weights. • 2 items • Updated 29 days ago • 177
view article Article The model that didn't exist, so you made it yourself ysharma, abidlabs • Oct 8 • 66
Why Self-Training Helps and Hurts: Denoising vs. Signal Forgetting Paper • 2602.14029 • Published Feb 15 • 1
The Problem Is the Problem: Towards Scalable Mathematical Discovery Paper • 2608.16977 • Published Aug 17 • 7
DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression Paper • 2609.19969 • Published Sep 17 • 208
Qwen-Image21-Doodle-in-LORA Collection Scribble-to-object edit LoRA for Qwen-Image 2.1: model, training pairs and demo, built with ML Intern. • 3 items • Updated Sep 29 • 1
view article Article tokenizers v1: encode, decode and scaling, measured +2 ArthurZ, sbrandeis, mcpotato, lysandre • Sep 21 • 97
view article Article Training a coding model to paint watercolours with TRL and OpenEnv sergiopaniego • Sep 3 • 82
view article Article Wire It, Run It, Deploy It: AI Workflows in Gradio ysharma, abidlabs • Aug 25 • 50
view article Article Shipping a Trillion Parameters With a Hub Bucket: Delta Weight Sync in TRL +6 aminediroHF, qgallouedec, kashif, lewtun, edbeeching, albertvillanova, lvwerra, sergiopaniego • May 27 • 50
view article Article State of Open Models: Summer 2026 Observations +1 AdinaY, multimodalart, irenesolaiman • Aug 14 • 230
view article Article Anatomy of a Frontier Lab Agent Intrusion: A Technical Timeline of the July 2026 Incident +2 hlarcher, XciD, raphael-gl, chris-rannou • Jul 27 • 518
view article Article Be Ready Before the Attack: A Practical Guide to Self-Hosting an Open Model for Cyber Defense jeffboudier • Jul 20 • 24