CARTE - Ile-de-France: leaderboard
Metric: Accuracy (%) on the 273 CARTE questions about Ile-de-France, 0-shot, five options per question (four answers and 'Je ne sais pas', which scores as wrong); questions generated by Gemini 3 Flash from human-selected French documents and filtered; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 27 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash | 96.4 |
| 2 | Mistral Nemo Instruct (2407) | 81 |
| 3 | Gemma 3 12B (IT) | 80.6 |
| 4 | Llama 3.1 8B Instruct | 78.4 |
| 5 | Qwen 3.5 9B | 77.7 |
| 6 | aya-expanse-8B | 75.1 |
| 7 | Mistral 7B Instruct (v0.3) | 74.4 |
| 8 | Mistral 7B Instruct (v0.2) | 68.5 |
| 9 | Llama 3.2 3B Instruct | 67.8 |
| 10 | Qwen 3.5 4B | 64.8 |
| 11 | Llama 3.2 1B Instruct | 53.8 |
| 12 | Lucie-7B | 47.6 |
| 13 | Luth-LFM2-1.2B | 46.9 |
| 14 | Mistral-7B-v0.1 | 30 |
| 15 | CroissantLLMChat-v0.1 | 24.2 |
Interactive version: theaggregate.ai/benchmark?slug=carte-ile-de-france · How It Works · Data refreshed daily, snapshot 2026-09-29.