AMA-Bench — leaderboard
Agent and memory benchmark spanning text-to-SQL, software, web, games, embodied AI, and open-world question answering tasks.
Metric: Average Score (%). Source: huggingface.co. Status: years away from saturation. 24 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.2 | 69.83 |
| 2 | GPT-5 Mini | 65.64 |
| 3 | Qwen 3 32B | 50.17 |
| 4 | Gemini 2.5 Flash | 49.22 |
| 5 | Qwen 3 14B | 45.24 |
| 6 | Qwen2.5-14B-Instruct-1M | 45.07 |
| 7 | Claude 3.5 Haiku | 43.12 |
| 8 | Qwen 3 8B | 39.89 |
Interactive version: theaggregate.ai/benchmark?slug=ama-bench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.