Ornith has released Ornith-1.5, an open model family with 3 different model variants: 397B MoE, 35B MoE, and 9B dense.
Ornith-1.5 has been build via end-to-end self-improvement. The model proposed new tasks, generated task-specific scaffolds, and produced solution rollouts for reinforcement learning. Ornith-1.5 performs at a level of Opus 4.8 and GLM 5.2 on most benchmarks.
Local testing time 👀
Aloha! 🌺Introducing Ornith-1.5, a family of open-source LLMs spanning 9B Dense, 35B MoE, and 397B MoE, trained with self-improving strategies. It achieves stat...