Tag
GPT5.6 Sol Ultra achieves 91.9% on TerminalBench coding benchmark, suggesting coding tasks are approaching solved.
This paper introduces RLVP (Reward the Outcome, Penalize the Path), a reinforcement learning method that uses a verifiable penalty for path violations and outcome reward to achieve near-zero constraint violations with high task success, improving sample efficiency in real-world agentic environments.
TerminalBench 2.1 is a benchmark suite derived from GPT‑5.6 Sol, Terra, and Luna models, likely used for evaluating AI performance on terminal-based tasks.