Skip to content
The Aggregate

Unified LLM rankings, daily

Models Task Explorer Subscriptions What's New Trends Benchmarks Builder Guesswork
How It Works Daily brief
The Aggregate
Loading data...

Nejumi 4 - BFCL - Relevance Detection: leaderboard

Metric: Accuracy (%). Source: nejumi.ai. 137 models tracked.

Top models

#ModelScore
1Llama 4 Maverick Instruct FP8100
2Grok 4.5 (High)100
3Grok 4.20 0309 (Reasoning)100
4qwen2.5-bakeneko-32B-instruct-v2100
5ELYZA-Shortcut-1.0-Qwen-32B100
6Gemma 3 12B (IT)94.44
7Qwen 3 Next 80B A3B Instruct94.44
8Qwen 3 235B A22B 2507 (Thinking)94.44
9Llama 3.1 405B Instruct FP894.44
10Llama-3.3-Swallow-70B-Instruct-v0.494.44
11GPT-4o Mini (2024-07-18)88.89
12Grok 488.89
13Qwen 3 30B A3B 2507 (Thinking)88.89
14DeepSeek V3.2 (Thinking)88.89
15DeepSeek V4 Pro (Max)88.89

Interactive version: theaggregate.ai/benchmark?slug=nejumi-4-bfcl-relevance-detection · How It Works · Data refreshed daily, snapshot 2026-09-19.

Built by Mikhail Doroshenko — AI researcher, co-author of Humanity’s Last Exam. Independent project; no affiliation with any AI lab.
A

Your Aggregate

Sign in to keep your work

Save custom rankings and get 50 free task analyses each day.

Continue with Google Continue with GitHub

No password to remember. Your public profile identifies saved work.

Your Aggregate

Account

  • Saved custom aggregate views
  • 50 free task analyses per day
Sign out