← Back to the wire

ornith-ai/Ornith-1.5-35B-A3B · Hugging Face

AnnouncementModelAug 19, 2026

ornith-ai has released Ornith-1.5-35B-A3B, a mid-size mixture-of-experts model that activates roughly 3B parameters per token, on Hugging Face. The model extends Ornith-1.0 with a self-improvement loop that jointly optimizes task generation, scaffold construction, and solution rollouts via reinforcement learning. The card claims it outperforms Qwen 3.6-35B on coding and agentic benchmarks and exceeds dense models like Gemma 4-31B on agentic coding. Usage instructions cover Transformers, vLLM, SGLang, and Docker.

Receipt № 14981 source · awaiting confirmation ◐

Evidence

1source· awaiting independent confirmation

No score is assigned. Sources and their independence are shown in the citation chain below.

Citation chain · 1 source

01highPRIMARY
Hugging FaceCompanyornith-aiCompanyOrnith-1.5-35B-A3BModel
Canonical: https://huggingface.co/ornith-ai/Ornith-1.5-35B-A3B