Smolagents - Overall (Tool Calling): leaderboard
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 6 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o | 53.21 |
| 2 | Qwen 2.5 Coder 32B Instruct | 40.25 |
| 3 | DeepSeek R1 Distill Qwen 32B | 34.83 |
| 4 | Claude 3.5 Sonnet | 18 |
| 5 | Llama 3.1 70B Instruct | 17.58 |
| 6 | Llama 3.1 8B Instruct | 10.08 |
Interactive version: theaggregate.ai/benchmark?slug=smolagents-overall-tool-calling · How It Works · Data refreshed daily, snapshot 2026-10-09.