LiveK12Bench - Biology: leaderboard
Final-answer accuracy (%) on the biology questions of LiveK12Bench (2026), a live benchmark of recent Chinese high-school examination papers in text, text-image and full-page formats, checked by a panel of LLM judges; higher is better.
Metric: Answer accuracy (%), pass@1 on the final answer, Biology questions, LiveK12Bench 2026-03 split of recent Chinese high-school exam papers, unified answer prompt, answers checked by a panel of three LLM judges that never grade their own model; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 12 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro | 90.7 | #145 |
| 2 | Claude Opus 4.6 | 88 | #113 |
| 3 | Gemini 3 Flash | 88 | #175 |
| 4 | Kimi K2.5 | 86.7 | #266 |
| 5 | Qwen 3 VL 32B | 86.7 | #622 |
| 6 | Qwen 3 VL 235B A22B | 85.5 | #485 |
| 7 | Claude Sonnet 4.6 | 84.9 | #173 |
| 8 | GLM-5 | 77.2 | #269 |
| 9 | Qwen 3 VL 8B | 75.3 | #917 |
| 10 | GPT-5 | 71.6 | #178 |
| 11 | GPT-5 Mini | 55.9 | #364 |
| 12 | GPT-4o | 44.1 | #751 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=livek12bench-biology · How It Works · Data refreshed daily, snapshot 2026-11-21.