Ai2's byte-level model Bolmo lands in Nature, releases new open checkpoints
allen_ai · x · 2026-10-07
Ai2's Bolmo, a family of fully open byte-level language models, is now published in Nature. Bolmo skips subword tokenization and processes raw bytes, improving handling of spelling, rare words, and diverse writing systems—bytes could also extend to images and audio.
- New checkpoints extend the approach beyond Olmo to other model families.
- Stage 1 checkpoints keep the original global model frozen while the byte-level components are trained, letting researchers experiment with new architectures without repeating the costly initial stage.
All weights are available on Hugging Face.
More from Models
- Anthropic Launches Claude Haiku 5.5, Costing ~75% Less Than Its Predecessor — rickasaurus · 2026-10-08
- Anthropic researcher's haiku hints Claude Haiku is jumping from 4.5 to 5.5 — edwinarbus · 2026-10-08
- Claude Haiku 5.5 spotted live in user accounts as rollout begins — rickasaurus · 2026-10-08
- Math professor grades OpenAI's 722 results: mostly B/C level, one D-level shock — khademinori · 2026-10-08
- Musk: Grok Bot requests will run on lightning-fast Grok 4.8 optimized for speed — XFreeze · 2026-10-08
- GitHub HydraFusion adds local model routing as Microsoft ships MAI Code 1.1 at 3-bit, 256K — BenBajarin · 2026-10-08