Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 9,511 prompts · See how Parse measures this
GTAgithub.com · Low evidenceGTA (General Tool Agent) is a benchmark and evaluation framework for General Tool Agents, featured at NeurIPS 2024. It evaluates both atomic tool use and long-horizon, open-ended workflows, focusing on what agents can accomplish in complete tasks rather than just predicting tool calls.
Parse Score