AI & ML interests
Post-training, model behavior and transparent evaluation. Research from alphaBell.
Recent Activity
View all activity
Articles
alphabell-research
published an article 4 days ago
Article
Ling 3.1 Flash 1019: same setup, new weight shards
alphabell
• alphabell-research
published an article 5 days ago
Article
Draw a MOSS layout once, then run it in ComfyUI
alphabell
• alphabell-research
published an article 5 days ago
Article
52 full-attention layers versus 16: mapping Nanbeige and Qwen
alphabell
• alphabell-research
published an article 8 days ago
Article
Five experiments inside Flint and Qwen Open
alphabell
• alphabell-research
published an article 9 days ago
Article
Draw MOSS-Image layouts without doing coordinate arithmetic
alphabell
• alphabell-research
published an article 9 days ago
Article
Three tiny bugs that broke Flint's games
alphabell
• alphabell-research
published an article 10 days ago
Article
alphaBell Qwen Open: 27B post-training, with the tradeoffs measured
alphabell
•