HumanEval
Résultats des évaluations de l’IA et performances des modèles
Type: percentage
Résultats enregistrés
82
Score moyen
82.66%
Plage des scores
39.70% - 92.90%
Modèle en tête
92.90% - Mistral Small 3.2
Scores au fil du temps
Scores individuels du benchmark tracés par date.
Résultats des modèles
| Organisation | Modèles | Date du résultat | Meilleur score | Informations | Déclaré par le fournisseur | Source |
|---|---|---|---|---|---|---|
| Mistral Small 3.2 | 20 juin 2025 | 92,90 % | Pass@5 | Oui | Source | |
| o1 mini | 12 sept. 2024 | 92,40 % | - | Oui | Source | |
| Mistral Large 2.1 | 18 nov. 2024 | 92 % | inferred version-family alias from mistral-large-2-2407 | Oui | Source | |
| Mistral OCR 2 | 22 mai 2025 | 92 % | inferred family alias from mistral-large-2-2407 (score=0.3100; benches=5) | Oui | Source | |
| Qwen 2.5 Coder 32B Instruct | - | 91,50 % | inferred high-confidence family alias from qwen2.5-vl-32b (score=0.4641; benches=28) | Oui | Source | |
| Qwen 2.5 VL 32B Instruct | 24 mars 2025 | 91,50 % | - | Oui | Source | |
| Qwen 2.5 VL 3B Instruct | - | 91,50 % | inferred high-confidence family alias from qwen2.5-vl-32b (score=0.4914; benches=28) | Oui | Source | |
| GPT 4o (2024-05-13) | 13 mai 2024 | 90,20 % | - | Oui | Source | |
| Granite 3.0 8B Instruct | - | 89,73 % | inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.4911; benches=14) | Oui | Source | |
| Granite 3.1 8B Instruct | - | 89,73 % | inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.4911; benches=14) | Oui | Source | |
| Granite 3.2 8B Instruct | - | 89,73 % | inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.4911; benches=14) | Oui | Source | |
| Granite 3.2 8B Instruct Preview | - | 89,73 % | inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.4687; benches=14) | Oui | Source | |
| Granite 3.3 2B Instruct | 16 avr. 2025 | 89,73 % | inferred family alias from granite-3.3-8b-instruct (score=0.3627; benches=14) | Oui | Source | |
| Granite 3.3 8B Instruct | 16 avr. 2025 | 89,73 % | - | Oui | Source | |
| Granite Guardian 3.0 8B | - | 89,73 % | inferred family alias from granite-3.3-8b-instruct (score=0.4062; benches=14) | Oui | Source | |
| Granite Guardian 3.1 8B | - | 89,73 % | inferred family alias from granite-3.3-8b-instruct (score=0.4062; benches=14) | Oui | Source | |
| Granite Guardian 3.3 8B | - | 89,73 % | inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.5071; benches=14) | Oui | Source | |
| Granite Speech 3.2 8B | - | 89,73 % | inferred family alias from granite-3.3-8b-instruct (score=0.4062; benches=14) | Oui | Source | |
| Granite Speech 3.3 8B | - | 89,73 % | inferred high-confidence family alias from granite-3.3-8b-instruct (score=0.5071; benches=14) | Oui | Source | |
| Gemini Diffusion | 20 mai 2025 | 89,60 % | Pass@1 | Oui | Source | |
| DeepSeek V2.5 (2024-09-05) | 05 sept. 2024 | 89 % | inferred alias from deepseek-v2.5 | Oui | Source | |
| DeepSeek V2.5 (2024-12-10) | 10 déc. 2024 | 89 % | inferred alias from deepseek-v2.5 | Oui | Source | |
| Granite 4.1 30B | 29 avr. 2026 | 88,41 % | pass@1 | Oui | Source | |
| LongCat Flash Chat | - | 88,41 % | inferred high-confidence family alias from longcat-flash-chat (score=0.4667; benches=16) | Oui | Source | |
| Llama 2 70B Chat | 20 juin 2023 | 88,40 % | inferred family alias from llama-3.3-70b-instruct (score=0.3129; benches=9) | Oui | Source |
