Jamba Reasoning 3B Collection AI21's top-performing reasoning model that packs leading scores on intelligence benchmarks and highly-efficient processing into a compact 3B build • 2 items • Updated 16 days ago • 5
LLaDA-8B-BGPO Collection Boundary-Guided Policy Optimization for Memory-Efficient RL of Diffusion Large Language Models • 4 items • Updated 13 days ago • 4