ScienceArena - INChO 2026: leaderboard
Metric: Score (out of 106): points awarded on the Indian National Chemistry Olympiad 2026, text-only question renderings; interleaved solving, one subpart at a time with earlier turns in context; open-ended physics and chemistry turns graded against the official solution and process-credit rubric by the authors' medalist-calibrated LLM-judge pipeline. Source: arxiv.org. Saturation forecast: Around December 2026. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 95.95 |
| 2 | Gemini 3.5 Flash | 92.25 |
| 3 | DeepSeek V4 Pro | 89.75 |
| 4 | Seed 2.0 Pro | 87.2 |
| 5 | MiMo-V2.5-Pro | 86 |
| 6 | GLM-5.1 | 85.75 |
| 7 | DeepSeek V4 Flash | 83.8 |
| 8 | GPT-5.4 | 83.5 |
| 9 | GPT-5.5 | 83 |
| 10 | Qwen 3.7 Max | 83 |
| 11 | MiniMax-M2.7 | 82.75 |
| 12 | Claude Opus 4.7 | 82.7 |
| 13 | Gemini 3 Flash | 81.5 |
| 14 | Qwen 3.6 Plus | 79.25 |
Interactive version: theaggregate.ai/benchmark?slug=sciencearena-incho-2026 · How It Works · Data refreshed daily, snapshot 2026-09-29.