ToolBench - Trip Booking — leaderboard

Metric: Task Score. Source: huggingface.co. 44 models tracked.

Top models

#ModelScore
1GPT-496.7
2text-davinci-00389.2
3CodeLlama-34B-Instruct-hf89.17
4CodeLlama-34B-hf88.33
5CodeLlama-34B-Python-hf85.83
6GPT-3.5 Turbo85.8
7CodeLlama-13B-Instruct-hf72.5
8Llama 2 70B71.67
9CodeLlama-13B-hf70.83
10CodeLlama-7B-Python-hf68.33
11LLaMA-65B65.8
12CodeLlama-7B-hf63.33
13CodeLlama-13B-Python-hf62.5
14CodeLlama-7B-Instruct-hf61.67
15Llama 2 13B53.33

Interactive version: theaggregate.ai/benchmark?slug=toolbench-trip-booking · How the rankings work · Data refreshed daily, snapshot 2026-07-22.