The original headline is: "Ornith 1.0: This is new class of self-improving model"

Short summary

Ornith 1 is a new family of open-weight models trained to generate both code solutions and task-specific harnesses using reinforcement learning, with the 397B variant approaching Opus 4.8 performance. The 9B model achieves 3-20× cost reduction vs. closed-source at comparable accuracy; scale (35B+) required for multi-step reasoning. Creator validates benchmarks on M2 Max and discusses reward-hacking defenses.

• •Ornith generates code solutions + task-specific harnesses in one loop using GRPO reinforcement learning
• •9B model achieves 3-20× cost reduction vs. closed-source with matching accuracy on some tasks
• •Creator validates on M2 Max hardware; 397B variant approaches Opus 4.8 performance

Generated with AI, which can make mistakes.