Smolagents - GAIA (Vanilla): leaderboard
Metric: Accuracy (%). Source: huggingface.co. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O1 | 18.75 |
| 2 | GPT-4.5 (Preview) | 15.63 |
| 3 | Qwen 2.5 72B Instruct | 6.25 |
| 4 | GPT-4o | 3.13 |
| 5 | Llama 3.3 70B Instruct | 3.13 |
| 6 | Qwen 2.5 7B Instruct | 3.13 |
| 7 | Claude 3.5 Sonnet | 3.13 |
| 8 | Qwen 2.5 Coder 32B Instruct | 3.13 |
| 9 | Llama 3.1 8B Instruct | 0 |
| 10 | Llama 3.2 3B Instruct | 0 |
Interactive version: theaggregate.ai/benchmark?slug=smolagents-gaia-vanilla · How It Works · Data refreshed daily, snapshot 2026-10-09.