Running on CPU Upgrade 1.26k 1.26k The Smol Training Playbook: The Secrets to Building World-Class LLMs 📝
MiniMax-M1 Collection MiniMax-M1, the world's first open-weight, large-scale hybrid-attention reasoning model. • 6 items • Updated 14 days ago • 114
view article Article 3+ Years of ML & Society at Hugging Face 🤗🤝🧑🤝🧑 By yjernite and 3 others • 6 days ago • 13
gpt-oss-safeguard Collection gpt-oss-safeguard-120b and gpt-oss-safeguard-20b are safety reasoning models built-upon gpt-oss • 2 items • Updated 6 days ago • 52
view article Article Cosmos Predict 2.5 & Transfer 2.5: Evolving the World Foundation Models for Physical AI By nvidia • 7 days ago • 17