← Benchmarks
Meilleurs modèles IA au benchmark Mathématiques
Classement des modèles IA par score au benchmark Mathématiques.
| Rang | Modèle | Fournisseur | Score | Jeu de données | Version | Date |
|---|---|---|---|---|---|---|
| 1 | Claude 3.5 Sonnet | Anthropic | 88.7 | mmlu | official-2024 | 2024-06-20 |
| 2 | GPT-4o | OpenAI | 88.7 | mmlu | official-2024 | 2024-05-13 |
| 3 | Llama 3.1 405B | Meta | 88.6 | mmlu | official-2024 | 2024-07-23 |
| 4 | DeepSeek V3 | DeepSeek | 88.5 | mmlu | paper-2024 | 2024-12-26 |
| 5 | Claude 3 Opus | Anthropic | 86.8 | mmlu | official-2024 | 2024-03-04 |
| 6 | Qwen2.5 72B | Alibaba | 86.1 | mmlu | official-2024 | 2024-09-19 |
| 7 | Llama 3.3 70B | Meta | 86 | mmlu | official-2024 | 2024-12-06 |
| 8 | Mistral Large 2 | Mistral | 84 | mmlu | official-2024 | 2024-07-24 |
| 9 | GPT-4o mini | OpenAI | 82 | mmlu | official-2024 | 2024-07-18 |
| 10 | Gemini 1.5 Pro | 81.9 | mmlu | techreport-2024 | 2024-05-14 | |
| 11 | GLM-4 | Zhipu | 81.5 | mmlu | official-2024 | 2024-01-16 |
| 12 | DeepSeek R1 | DeepSeek | 79.8 | aime | paper-2025 | 2025-01-20 |
| 13 | Gemini 1.5 Flash | 78.9 | mmlu | techreport-2024 | 2024-05-14 | |
| 14 | Claude 3 Haiku | Anthropic | 75.1 | mmlu | official-2024 | 2024-03-04 |
Source
- Jeu de données
- mmlu
- Version
- official-2024
- Date
- 2024-06-20
Benchmark
Related Resources
Use Cases