alphaBell

company

https://alphabell.com

alphabell-research

AI & ML interests

Post-training, model behavior and transparent evaluation. Research from alphaBell.

Recent Activity

alphabell-research new activity 2 minutes ago

alphabell/README:Feature five measured experiments and matched checkpoint downloads

alphabell-research published an article 4 minutes ago

We tried to catch two models repeating an outdated fact. Both passed.

alphabell-research new activity 7 minutes ago

alphabell/Model-Observatory:Add five measured experiments and public CPU reproduction

View all activity

Articles

We tried to catch two models repeating an outdated fact. Both passed.

Draw MOSS-Image layouts without doing coordinate arithmetic

about 5 hours ago

Three tiny bugs that broke Flint's games

about 10 hours ago

alphaBell Qwen Open: 27B post-training, with the tradeoffs measured

View all articles

alphabell 's models 3

alphabell/alphaBell-Qwen3.8-27B-Open-GGUF

Text Generation • 27B • Updated 9 minutes ago

alphabell/Flint-Matched-GGUF

Text Generation • 40B • Updated 9 minutes ago

alphabell/alphaBell-Qwen3.8-27B-Open-FP8

Image-Text-to-Text • 28B • Updated 1 day ago • 27