Skip to content

Benchmark

BFCL Relevance

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
0.25

Score distribution

77.3%
Mean
81.3%
Median
18.5%
Std dev
50%
10th percentile
93.8%
90th percentile
0%
Min
0%100%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.25 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 Medium 3· Berkeley Function-Calling Leaderboard data75%
52Nova 2 Litefc· Berkeley Function-Calling Leaderboard data75%
53Qwen3 0.6Bfc· Berkeley Function-Calling Leaderboard data75%
54Qwen3 0.6BPrompt· Berkeley Function-Calling Leaderboard data75%
55Qwen3 8BPrompt· Berkeley Function-Calling Leaderboard data75%
56Claude Opus 4.5Prompt· Berkeley Function-Calling Leaderboard data68.8%
57Claude Sonnet 4.5fc· Berkeley Function-Calling Leaderboard data68.8%
58Command Afc· Berkeley Function-Calling Leaderboard data68.8%
59command-r7bfc· Berkeley Function-Calling Leaderboard data68.8%
60Gemini 3 Pro PreviewPrompt· Berkeley Function-Calling Leaderboard data68.8%
61GPT-4.1 nanoPrompt· Berkeley Function-Calling Leaderboard data68.8%
62Granite 3.1 8B Instructfc· Berkeley Function-Calling Leaderboard data68.8%
63Claude Haiku 4.5fc· Berkeley Function-Calling Leaderboard data62.5%
64Claude Opus 4.5fc· Berkeley Function-Calling Leaderboard data62.5%
65Gemini 2.5 FlashPrompt· Berkeley Function-Calling Leaderboard data62.5%
66GPT-5 minifc· Berkeley Function-Calling Leaderboard data62.5%
67Mistral Medium 3fc· Berkeley Function-Calling Leaderboard data62.5%
68Gemini 2.5 Flash-LitePrompt· Berkeley Function-Calling Leaderboard data50%
69Phi 4Prompt· Berkeley Function-Calling Leaderboard data50%
70Gemini 2.5 Flash-Litefc· Berkeley Function-Calling Leaderboard data43.8%
71Llama 3.2 1B Instructfc· Berkeley Function-Calling Leaderboard data43.8%
72Claude Sonnet 4.5Prompt· Berkeley Function-Calling Leaderboard data37.5%
73DeepSeek V3.2 Expfc· Berkeley Function-Calling Leaderboard data37.5%
74Gemma 3 1b itPrompt· Berkeley Function-Calling Leaderboard data37.5%
75Claude Haiku 4.5Prompt· Berkeley Function-Calling Leaderboard data31.3%
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)