BAGEL - Xeno-canto Harmonic Structure: leaderboard

BAGEL Xeno-canto questions on whether a vocalization is harmonic or broadband, 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 427 Xeno-canto questions on harmonic structure versus broadband noise 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 2032. 13 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1Llama 3.1 8B Instruct46.4#4655
2Qwen 3 4BNon-reasoning40.3#2379
3Qwen 3 32BNon-reasoning39.3#983
4Gemma 3 27B (IT)38.9#1341
5Mistral 7B Instruct (v0.3)37.7#6759
6GPT-5.435.1#138
7Qwen 3 14BNon-reasoning32.3#1284
8gemma-7B32.1#7115
9Qwen 3 8BNon-reasoning29.7#1890
10Claude Opus 4.628.6#114
11Phi-428.6#1951
12SmolLM2-360M26.9#9572
13Qwen 3 0.6BNon-reasoning22.5#8357

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=bagel-xeno-canto-harmonic-structure · How It Works · Data refreshed daily, snapshot 2026-11-22.