J.O.S.I.E.v1 Collection The first and only LLM family, trained completely from scratch on apple silicon. • 0 items • Updated 8 days ago • 1
Olmo-3 Collection Ai2's Olmo 3 model family of instruction and reasoning models. • 32 items • Updated 13 days ago • 4
Olmo 3 Pre-training Collection All artifacts related to Olmo 3 pre-training • 10 items • Updated 9 days ago • 27
Olmo 3 Post-training Collection All artifacts for post-training Olmo 3. Datasets follow the model that resulted from training on them. • 32 items • Updated 7 days ago • 38
Tiny Model, Big Logic: Diversity-Driven Optimization Elicits Large-Model Reasoning Ability in VibeThinker-1.5B Paper • 2511.06221 • Published 30 days ago • 128
view article Article Mitigating False Negatives in Multiple Negatives Ranking Loss for Retriever Training May 25 • 24
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 Oct 8 • 5