Ornith 1.5 (35B-A3B, Q4)

Ornith AI's Ornith 1.5 at 35B total parameters with roughly 3B active per token — a self-improving, agentic-coding model trained for software engineering, tool use, and reasoning. The 4-bit build is the practical pick at this size, keeping fast MoE decode in a much smaller footprint than Q8, with native tool calling and a 256K context window. MIT licensed; best suited to machines with 32GB+ usable memory.

Gezel downloads this model the first time a gezel needs it, checks the file against its published hash, and applies tuned settings for it.

The model’s own page · Every model · Model scorecard