Eșecuri AI BENCHY
Eșecuri Timp expirat
Vezi ce modele AI se lovesc cel mai des de Timp expirat, ca să identifici riscurile de fiabilitate înainte să alegi.
| Rang | Model | Companie | Număr de Timp expirat | Scor | Teste corecte | Timp de răspuns (mediu) |
|---|---|---|---|---|---|---|
| #142 | Qwen3.5-9B medium | Qwen | 10 | 4.3 | 3/19 | 80.1s |
| #58 | Seed-2.0-Mini medium | Bytedance Seed | 4 | 7.2 | 11/19 | 68.9s |
| #59 | Qwen3.5-35B-A3B medium | Qwen | 4 | 7.2 | 10/19 | 51.5s |
| #108 | MiniMax M2.5 medium | Minimax | 4 | 5.5 | 5/19 | 43.4s |
| #45 | Qwen3.5-Flash medium | Qwen | 3 | 7.6 | 11/19 | 65.8s |
| #12 | Qwen3.5 Plus 2026-02-15 medium | Qwen | 2 | 8.2 | 14/19 | 51.3s |
| #31 | Qwen3.5-122B-A10B medium | Qwen | 2 | 7.9 | 13/19 | 32.5s |
| #38 | Gemma 4 26B A4B medium | 2 | 7.7 | 13/19 | 33.7s | |
| #42 | Kimi K2.6 medium | Moonshot AI | 2 | 7.6 | 12/19 | 49.9s |
| #47 | GLM 5.1 medium | Z.ai | 2 | 7.6 | 12/19 | 24.4s |
| #61 | DeepSeek V3.2 medium | DeepSeek | 2 | 7.2 | 11/19 | 46.1s |
| #71 | Kimi K2.5 medium | Moonshot AI | 2 | 6.8 | 9/19 | 73.4s |
| #73 | Hunter Alpha medium | OpenRouter | 2 | 6.7 | 8/18 | 10.3s |
| #125 | MiniMax M2.7 medium | Minimax | 2 | 5.1 | 4/19 | 30.6s |
| #3 | Claude Opus 4.7 medium | Anthropic | 1 | 8.9 | 16/19 | 3.46s |