OMIBench - Chemistry: leaderboard
OMIBench (2026) tests vision-language models on Olympiad-level science and mathematics problems whose evidence is spread across several images. This board is the GPTScore in percent on the chemistry problems; higher is better.
Metric: GPTScore (%, binarized LLM-judge equivalence of the final answer with the gold answer, micro-averaged) on the 217 chemistry problems of OMIBench (Olympiad-level multi-image problems from biology, chemistry, mathematics and physics competitions), zero-shot; higher is better. Source: arxiv.org. Saturation forecast: Around April 2028. 37 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 VL 235B A22B (Thinking) | 33.18 | #459 |
| 2 | O4 Mini | 32.41 | #368 |
| 3 | Qwen 3 VL 30B A3B (Thinking) | 32.26 | #719 |
| 4 | GPT-5 | 29.03 | #178 |
| 5 | Gemini 3 Pro (Preview) | 25.35 | #119 |
| 6 | InternVL3.5-241B-A28B | 25.35 | #733 |
| 7 | Qwen 3 VL 32B (Thinking) | 24.88 | #573 |
| 8 | GPT-5 Mini | 24.42 | #364 |
| 9 | Gemini 2.5 Pro | 23.96 | #276 |
| 10 | GPT-4o | 22.58 | #751 |
| 11 | Qwen 3 VL 235B A22B Instruct | 22.58 | #532 |
| 12 | InternVL3.5-38B | 22.12 | #1021 |
| 13 | GPT-4o Mini | 21.66 | #1667 |
| 14 | InternVL3.5-14B | 21.2 | #1504 |
| 15 | InternVL3-78B | 20.74 | #759 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=omibench-chemistry · How It Works · Data refreshed daily, snapshot 2026-11-21.