Eșecuri AI BENCHY
Eșecuri Formatare suplimentară
Vezi ce modele AI se lovesc cel mai des de Formatare suplimentară, ca să identifici riscurile de fiabilitate înainte să alegi. Sortează după: Timp de răspuns (mediu) ↑.
| Rang | Model | Companie | Număr de Formatare suplimentară | Scor | Teste corecte | Timp de răspuns (mediu) |
|---|---|---|---|---|---|---|
| #163 | Granite 4.1 8B none | IBM Granite | 1 | 4.0 | 2/21 | 728ms |
| #127 | Grok 4.20 none | X AI | 1 | 5.4 | 6/18 | 1.11s |
| #143 | MiMo-V2.5 none | Xiaomi | 1 | 4.9 | 5/21 | 2.20s |
| #101 | Mimo V2 Omni none | Xiaomi | 1 | 6.0 | 8/21 | 2.44s |
| #152 | MiMo-V2-Flash none | Xiaomi | 1 | 4.6 | 4/21 | 2.76s |
| #68 | Claude Opus 4.8 none | Anthropic | 3 | 7.0 | 12/21 | 3.47s |
| #77 | Claude Sonnet 4.6 none | Anthropic | 4 | 6.8 | 11/21 | 5.04s |
| #140 | Qwen3 Coder Next none | Qwen | 1 | 4.9 | 5/21 | 8.62s |
| #84 | Grok 4.20 Multi Agent Beta medium | X AI | 2 | 6.6 | 8/18 | 9.69s |
| #121 | Owl Alpha none | Openrouter | 1 | 5.5 | 7/21 | 9.88s |
| #79 | Hunter Alpha medium | OpenRouter | 1 | 6.7 | 8/18 | 10.3s |
| #113 | DeepSeek V4 Pro none | DeepSeek | 1 | 5.7 | 7/21 | 12.4s |
| #156 | Hy3 preview none | Tencent | 1 | 4.4 | 4/21 | 12.9s |
| #133 | DeepSeek V3.2 none | DeepSeek | 2 | 5.2 | 6/21 | 13.8s |
| #52 | Claude Sonnet 4.6 medium | Anthropic | 3 | 7.4 | 13/21 | 17.1s |