Konkur 1404 - Foreign Language: leaderboard
The Foreign Language stream of Konkur 1404 — Persian-language multiple-choice questions from the foreign-language exam of Iran's 1404 (2025) national university entrance exam; text-only accuracy (%).
Metric: Accuracy (%, text-only). Source: huggingface.co. Saturation forecast: Not forecast. 23 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Ministral 3 14B | 44.23 | #650 |
| 2 | gpt-oss-safeguard-20B | 42.31 | #314 |
| 3 | Qwen 3 VL 235B A22B Instruct | 40.38 | #273 |
| 4 | Kimi K2 0905 | 40.38 | #288 |
| 5 | Gemma 3 27B (IT) | 39.42 | #523 |
| 6 | Gemma 3 4B (IT) | 38.46 | #992 |
| 7 | GPT-OSS-120B | 38.46 | #341 |
| 8 | Llama 3.1 70B Instruct | 37.5 | #561 |
| 9 | Qwen 3 Next 80B A3B Instruct | 37.5 | #368 |
| 10 | Qwen 3 VL 30B A3B Instruct | 37.5 | #384 |
| 11 | Devstral Small 2 | 37.5 | #495 |
| 12 | Llama 4 Maverick | 36.54 | #464 |
| 13 | Mistral Nemo | 35.58 | #1085 |
| 14 | Mixtral 8x22B Instruct | 34.62 | #728 |
| 15 | Gemma 3n E4B (IT) | 34.62 | #749 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=konkur-1404-foreign-language · How It Works · Data refreshed daily, snapshot 2026-10-20.