Souper-Model: How Simple Arithmetic Unlocks State-of-the-Art LLM Performance Paper • 2511.13254 • Published Nov 17, 2025 • 136
TiDAR: Think in Diffusion, Talk in Autoregression Paper • 2511.08923 • Published Nov 12, 2025 • 122
Running on CPU Upgrade Featured 2.84k The Smol Training Playbook 📚 2.84k The secrets to building world-class LLMs
Cerebras REAP Collection Sparse MoE models compressed using REAP (Router-weighted Expert Activation Pruning) method • 22 items • Updated 1 day ago • 82
cerebras/GLM-4.5-Air-REAP-82B-A12B Text Generation • 82B • Updated Oct 21, 2025 • 15.4k • 108