Categorie AI BENCHY
Clasament Trucuri anti-AI
Vezi ce modele AI se descurcă cel mai bine la Trucuri anti-AI, care rămân fiabile și unde apar cele mai mari diferențe. Sortează după: Metrică ↑.
| Rang | Model | Companie | Scor Trucuri anti-AI | Scor | Teste corecte | Timp de răspuns (mediu) |
|---|---|---|---|---|---|---|
| #127 | Grok 4.20 none | X AI | 4.8 | 5.4 | 1/4 | 501ms |
| #131 | Qwen3.5-122B-A10B none | Qwen | 4.8 | 5.3 | 1/4 | 1.59s |
| #147 | GPT-4o-mini none | OpenAI | 4.8 | 4.8 | 1/4 | 1.34s |
| #156 | Hy3 preview none | Tencent | 4.8 | 4.4 | 1/4 | 11.1s |
| #162 | Nemotron 3 Nano Omni 30b A3b Reasoning none | NVIDIA | 4.8 | 4.1 | 1/4 | 584ms |
| #141 | Nemotron 3 Super none | NVIDIA | 4.8 | 4.9 | 1/4 | 4.46s |
| #163 | Granite 4.1 8B none | IBM Granite | 4.9 | 4.0 | 1/4 | 844ms |
| #161 | Qwen3.5-9B medium | Qwen | 5.1 | 4.2 | 1/4 | 34.4s |
| #74 | Qwen3.6 Max Preview none | Qwen | 5.2 | 6.9 | 1/4 | 2.63s |
| #122 | GLM 4.7 Flash none | Z.ai | 5.2 | 5.5 | 1/4 | 5.51s |
| #67 | MiniMax M3 medium | Minimax | 5.5 | 7.1 | 1/4 | 14.9s |
| #132 | Mistral Small 4 medium | Mistral | 5.6 | 5.3 | 1/4 | 2.67s |
| #69 | Claude Opus 4.6 medium | Anthropic | 6.4 | 7.0 | 2/4 | 7.45s |
| #82 | Hy3 preview high | Tencent | 6.4 | 6.6 | 2/4 | 15.1s |
| #103 | DeepSeek V4 Pro high | DeepSeek | 6.4 | 6.0 | 2/4 | 16.5s |