AgentBoard — leaderboard
AgentBoard: Evaluates autonomous agent performance on multi-step tasks requiring planning, state tracking, tool use, and recovery.
Metric: Progress Rate (self-reported). Source: benchmarklist.com. Status: saturation imminent. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4 | 70 |
| 2 | Claude 2 | 48.9 |
| 3 | GPT-3.5 Turbo | 41.4 |
| 4 | text-davinci-003 | 37.2 |
| 5 | Mistral 7B | 24.6 |
| 6 | Llama 2 70B | 23.8 |
| 7 | Llama 2 13B | 18.9 |
Interactive version: theaggregate.ai/benchmark?slug=agentboard · How the rankings work · Data refreshed daily, snapshot 2026-07-22.