BAGEL - Xeno-canto Dominant Frequency: leaderboard

BAGEL Xeno-canto questions on the dominant frequency range of a species song or call, answered from text alone. Four-option closed-book questions written by GPT-4o-mini from source material that is withheld at test time; accuracy in percent, higher is better.

Metric: Accuracy (%, times 100) on the 2,747 Xeno-canto questions on a species dominant frequency range of BAGEL, closed-book four-option single-answer multiple choice (the source passage is withheld), one unified prompt, greedy decoding with seed 0, a single run per model; questions written by GPT-4o-mini from Wikipedia species articles, GloBI interaction records, bioRxiv preprints and Xeno-canto recordings, options shuffled; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 13 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1GPT-5.471.1#135
2Claude Opus 4.667.1#113
3Llama 3.1 8B Instruct58.4#4600
4Gemma 3 27B (IT)52.1#1324
5Phi-444.3#1931
6Qwen 3 14BNon-reasoning44.2#1269
7Qwen 3 8BNon-reasoning42.7#1873
8Qwen 3 32BNon-reasoning40.3#972
9Qwen 3 4BNon-reasoning37.5#2348
10Mistral 7B Instruct (v0.3)27.1#6750
11gemma-7B27#7121
12SmolLM2-360M24.8#9570
13Qwen 3 0.6BNon-reasoning23.7#8350

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=bagel-xeno-canto-dominant-frequency · How It Works · Data refreshed daily, snapshot 2026-11-21.