AI BENCHY श्रेणी
डोमेन-विशिष्ट क्रमवारी
डोमेन-विशिष्ट मध्ये कोणती AI मॉडेल्स सर्वोत्तम काम करतात, कोणती विश्वासार्ह राहतात आणि सर्वात मोठी दरी कुठे दिसते ते पाहा. क्रम लावा: बरोबर चाचण्या ↑.
| क्रमांक | मॉडेल | कंपनी | डोमेन-विशिष्ट स्कोअर | स्कोअर | बरोबर चाचण्या | प्रतिसाद वेळ (सरासरी) |
|---|---|---|---|---|---|---|
| #114 | Qwen3.5 Plus 2026-04-20 none | Qwen | 5.3 | 5.7 | 1/3 | 4.43s |
| #116 | Hunter Alpha none | OpenRouter | 5.3 | 5.7 | 1/3 | 2.33s |
| #120 | Mimo V2 PRO none | Xiaomi | 5.3 | 5.6 | 1/3 | 1.78s |
| #121 | Owl Alpha none | Openrouter | 5.3 | 5.5 | 1/3 | 3.00s |
| #123 | MiMo-V2.5-Pro none | Xiaomi | 5.3 | 5.5 | 1/3 | 877ms |
| #124 | Kimi K2.6 none | Moonshot AI | 5.3 | 5.5 | 1/3 | 1.48s |
| #125 | GPT-5.4 none | OpenAI | 5.3 | 5.5 | 1/3 | 1.07s |
| #128 | Qwen3.6 Flash none | Qwen | 5.3 | 5.4 | 1/3 | 1.11s |
| #131 | Qwen3.5-122B-A10B none | Qwen | 5.3 | 5.3 | 1/3 | 465ms |
| #132 | Mistral Small 4 medium | Mistral | 5.3 | 5.3 | 1/3 | 6.11s |
| #134 | GLM 5 Turbo none | Z.ai | 5.3 | 5.2 | 1/3 | 1.97s |
| #135 | Kimi K2.5 none | Moonshot AI | 5.3 | 5.2 | 1/3 | 4.38s |
| #139 | DeepSeek V4 Flash none | DeepSeek | 5.3 | 5.0 | 1/3 | 19.7s |
| #140 | Qwen3 Coder Next none | Qwen | 5.3 | 4.9 | 1/3 | 962ms |
| #142 | Mistral Small 4 none | Mistral | 5.3 | 4.9 | 1/3 | 367ms |