ToolSandbox — leaderboard
ToolSandbox: Evaluates tool calling, API use, function selection, structured arguments, and multi-step tool workflows.
Metric: Avg Score (self-reported). Source: benchmarklist.com. Status: saturation imminent. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o | 73 |
| 2 | Claude 3 Opus | 69.2 |
| 3 | GPT-3.5 Turbo | 65.6 |
| 4 | GPT-4 | 64.3 |
| 5 | Claude 3 Sonnet (20240229) | 63.8 |
| 6 | Gemini 1.5 Pro (001) | 60.4 |
| 7 | Claude 3 Haiku | 54.9 |
| 8 | Gemini 1.0 Pro | 38.1 |
| 9 | Hermes-2-Pro-Mistral-7B | 31.4 |
| 10 | Mistral 7B Instruct (v0.3) | 29.8 |
| 11 | c4ai-command-r-v01 | 26.2 |
| 12 | c4ai-command-r-plus | 24.7 |
Interactive version: theaggregate.ai/benchmark?slug=toolsandbox · How the rankings work · Data refreshed daily, snapshot 2026-07-22.