AgentBoard: leaderboard
AgentBoard: Evaluates autonomous agent performance on multi-step tasks requiring planning, state tracking, tool use, and recovery.
Metric: Progress Rate (imported). Source: github.com. Source: benchmarklist.com. Saturation forecast: Estimated already saturated. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4 | 70 | #438 |
| 2 | Claude 2 | 48.9 | #761 |
| 3 | GPT-3.5 Turbo | 41.4 | #865 |
| 4 | text-davinci-003 | 37.2 | #1146 |
| 5 | Mistral 7B | 24.6 | #1397 |
| 6 | Llama 2 70B | 23.8 | #1201 |
| 7 | Llama 2 13B | 18.9 | #1436 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=agentboard · How It Works · Data refreshed daily, snapshot 2026-10-20.