Ornith-1.0 Open-Source Coding Models Released
s_mohinii · x · 2026-07-13
- The open-source code model family Ornith-1.0 has been released, targeting agentic coding scenarios.
- It comes in various sizes: 9B Dense, 31B Dense, 35B MoE, and 397B MoE.
- The team claims it sets new highs on coding benchmarks among open-source models of similar scale, including:
- Terminal-Bench 2.1: 77.5
- SWE-Bench: verified 82.4, pro 62.2, Multilingual 78.9
- NL2Repo: 48.2
- SWE Atlas: QnA 41.2, RF 42.6, TW 39.1
- ClawEval: 77.1
- The models are continually trained based on gemma4 and qwen3.5, utilizing a self-improvement training strategy: they use reinforcement learning not only to generate problem-solving trajectories but also to create specific scaffolds that drive tasks.
More from coding & agent
- Cognition's SWE-2 uses a KKT duality argument in RL to shift the effort Pareto curve — YouJiacheng · 2026-09-11
- First-ever Three.js Conference lands in Paris, with a panel on AI-shortened design workflows — OdinLovis · 2026-09-11
- Data engineering, not agent frameworks, is the real bottleneck for enterprise AI agents — dhruv2038 · 2026-09-11
- RTK Terminal Compression Cuts Tokens but Leaves Your AI Coding Bill Unchanged — Bartaseth · 2026-09-11
- GPT-6 Astra beats Factorio with enemies in 44 in-game hours at ~$4,500 API cost — liminal_bardo · 2026-09-11
- Investment Analyst Asks How to Build a Claude-Based Diligence Agent Stack — Careless_Tie2286 · 2026-09-11