Skip to content

Benchmark

BFCL Live

Agentic benchmarks measure how well models carry out tasks that take several steps. Scores are percentages.

Models measured
49
Scores recorded
76
Sources
1
IQ weight
1.40

Score distribution

68.9%
Mean
74%
Median
15.2%
Std dev
54.9%
10th percentile
80.5%
90th percentile
0%
Min
0%83.1%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.40 for chat models, measured over 49 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
51Mistral LargePrompt· Berkeley Function-Calling Leaderboard data68.1%
52Mistral Medium 3fc· Berkeley Function-Calling Leaderboard data68%
53GPT 5.2Prompt· Berkeley Function-Calling Leaderboard data67.1%
54Nova Microfc· Berkeley Function-Calling Leaderboard data66.3%
55OpenAI o3fc· Berkeley Function-Calling Leaderboard data66.2%
56o4 Minifc· Berkeley Function-Calling Leaderboard data66.1%
57Mistral Medium 3· Berkeley Function-Calling Leaderboard data66%
58Gemini 2.5 Flash-Litefc· Berkeley Function-Calling Leaderboard data65.8%
59GPT-5 miniPrompt· Berkeley Function-Calling Leaderboard data62.6%
60Gemma 3 4B ITPrompt· Berkeley Function-Calling Leaderboard data60.8%
61GPT-4.1 nanofc· Berkeley Function-Calling Leaderboard data60.8%
62Phi 4Prompt· Berkeley Function-Calling Leaderboard data60.7%
63Granite 3.1 8B Instructfc· Berkeley Function-Calling Leaderboard data60.3%
64GPT-5 nanofc· Berkeley Function-Calling Leaderboard data59.4%
65GPT-5 minifc· Berkeley Function-Calling Leaderboard data58.6%
66Llama 3.2 3Bfc· Berkeley Function-Calling Leaderboard data58.3%
67Qwen3 0.6Bfc· Berkeley Function-Calling Leaderboard data56.6%
68Gemini 2.5 Flash-LitePrompt· Berkeley Function-Calling Leaderboard data54.9%
69DeepSeek V3.2 Expfc· Berkeley Function-Calling Leaderboard data53.7%
70Claude Haiku 4.5Prompt· Berkeley Function-Calling Leaderboard data52.5%
71GPT-4.1 nanoPrompt· Berkeley Function-Calling Leaderboard data50.3%
72Qwen3 0.6BPrompt· Berkeley Function-Calling Leaderboard data49.4%
73Claude Sonnet 4.5Prompt· Berkeley Function-Calling Leaderboard data46.6%
74Gemma 3 1b itPrompt· Berkeley Function-Calling Leaderboard data11.8%
75Llama 3.2 1B Instructfc· Berkeley Function-Calling Leaderboard data11.8%
76Llama 3.1 Nemotron Ultra 253B v1fc· Berkeley Function-Calling Leaderboard data0%

Showing 51–76 of 76 results.

Data credits: Berkeley Function-Calling Leaderboard (Gorilla, UC Berkeley), Apache-2.0 (licence)