BAGEL - Wikipedia Habitat: leaderboard

BAGEL Wikipedia questions on the habitat a species lives in. 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 234 Wikipedia habitat questions 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: Estimated already saturated. 13 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1Claude Opus 4.698.3#113
2GPT-5.497#135
3Qwen 3 32BNon-reasoning93.6#972
4Phi-493.2#1931
5Gemma 3 27B (IT)92.3#1324
6Qwen 3 14BNon-reasoning92.3#1269
7Qwen 3 8BNon-reasoning89.7#1873
8Llama 3.1 8B Instruct85.9#4600
9gemma-7B84.6#7121
10Mistral 7B Instruct (v0.3)83.8#6750
11Qwen 3 4BNon-reasoning81.2#2348
12Qwen 3 0.6BNon-reasoning25.6#8350
13SmolLM2-360M22.2#9570

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

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