Konkur 1404 - Experimental Sciences — leaderboard
Metric: Accuracy (%, text-only). Source: huggingface.co. 18 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3 VL 30B A3B Instruct | 50.31 |
| 2 | Qwen 3 Next 80B A3B Instruct | 46.58 |
| 3 | Qwen 3 VL 235B A22B Instruct | 46.58 |
| 4 | Kimi K2 0905 | 44.72 |
| 5 | Gemma 3 27B (IT) | 39.75 |
| 6 | Ministral 3 14B | 39.75 |
| 7 | Devstral Small 2 | 39.13 |
| 8 | Llama 4 Maverick | 37.27 |
| 9 | Mixtral 8x22B Instruct | 36.65 |
| 10 | command-r7B-12-2024 | 34.78 |
| 11 | Llama 3.1 70B Instruct | 34.16 |
| 12 | MiMo-V2-Flash | 33.54 |
| 13 | Gemma 3 4B (IT) | 32.3 |
| 14 | Llama 3.1 8B Instruct | 31.06 |
| 15 | Gemma 3n E4B (IT) | 28.57 |
Interactive version: aibenchmarks.dev/benchmark?slug=konkur-1404-experimental-sciences · How the rankings work · Data refreshed daily, snapshot 2026-07-12.