BFCL V4 - Relevance Detection: leaderboard

Metric: Accuracy (%). Source: gorilla.cs.berkeley.edu. 109 models tracked.

Top models

#ModelScore
1Llama 3.3 70B Instruct100
2Gemma 3 4B (IT)100
3GPT-4.1100
4Llama 4 Scout Instruct100
5Llama 4 Maverick Instruct FP8100
6Falcon3-7B-Instruct100
7Nanbeige3.5-Pro (Thinking)100
8Llama 3.1 8B Instruct93.75
9Gemma 3 12B (IT)93.75
10GPT-4.1 Nano93.75
11GPT-5 Nano93.75
12GPT-5 Mini93.75
13Qwen 3 8B93.75
14Qwen 3 30B A3B 2507 Instruct93.75
15O3 (2025-04-16)93.75

Interactive version: theaggregate.ai/benchmark?slug=bfcl-v4-relevance-detection · How It Works · Data refreshed daily, snapshot 2026-09-19.