A verifier that scores agent progress, not just final answers
Most agent evaluators wait for the final answer. This open-source framework scores each step instead, using fine-grained criteria and the model’s full score-token distribution to guide selection, progress tracking, and reinforcement learning. Its Terminal-Bench 2.1 example lifts Best-of-3 from 79.4% Pass@1 to 86.5%, even when the verifier judges rollouts from its own model.