ToolBench - Google Sheets — leaderboard
Metric: Task Score. Source: huggingface.co. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | CodeLlama-34B-hf | 64.29 |
| 2 | GPT-4 | 62.9 |
| 3 | text-davinci-003 | 62.9 |
| 4 | CodeLlama-34B-Instruct-hf | 61.11 |
| 5 | Llama 2 70B | 58.57 |
| 6 | CodeLlama-34B-Python-hf | 55.87 |
| 7 | GPT-3.5 Turbo | 51.4 |
| 8 | CodeLlama-13B-hf | 51.26 |
| 9 | CodeLlama-13B-Python-hf | 50.79 |
| 10 | CodeLlama-7B-Python-hf | 49.13 |
| 11 | CodeLlama-13B-Instruct-hf | 48.97 |
| 12 | starcoder | 48 |
| 13 | CodeLlama-7B-Instruct-hf | 41.27 |
| 14 | CodeLlama-7B-hf | 38.08 |
| 15 | LLaMA-30B | 37.1 |
Interactive version: theaggregate.ai/benchmark?slug=toolbench-google-sheets · How the rankings work · Data refreshed daily, snapshot 2026-07-22.