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.
| क्रमांक | मॉडेल | कंपनी | डोमेन-विशिष्ट स्कोअर | स्कोअर | बरोबर चाचण्या | प्रतिसाद वेळ (सरासरी) |
|---|---|---|---|---|---|---|
| #108 | Qwen3.5-Flash none | Qwen | 7.7 | 5.8 | 2/3 | 905ms |
| #117 | Qwen3.5-35B-A3B none | Qwen | 7.7 | 5.6 | 2/3 | 485ms |
| #118 | Qwen3.6 27B none | Qwen | 7.7 | 5.6 | 2/3 | 3.03s |
| #122 | GLM 4.7 Flash none | Z.ai | 7.7 | 5.5 | 2/3 | 744ms |
| #2 | Gemini 3.5 Flash high | 7.6 | 9.6 | 2/3 | 14.1s | |
| #20 | Gemini 3.5 Flash none | 7.6 | 8.1 | 2/3 | 10.6s | |
| #5 | Qwen3.7 Max medium | Qwen | 5.9 | 9.1 | 1/3 | 24.9s |
| #15 | GPT-5.3-Codex medium | OpenAI | 5.9 | 8.4 | 1/3 | 64.3s |
| #19 | Seed-2.0-Lite medium | Bytedance Seed | 5.9 | 8.2 | 1/3 | 88.7s |
| #28 | Gemini 2.5 Flash medium | 5.9 | 7.8 | 1/3 | 37.3s | |
| #42 | GPT-5.2 medium | OpenAI | 5.9 | 7.5 | 1/3 | 77.8s |
| #64 | MiMo-V2-Flash medium | Xiaomi | 5.9 | 7.2 | 1/3 | 96.0s |
| #70 | GPT-5.4 Nano medium | OpenAI | 5.9 | 7.0 | 1/3 | 38.2s |
| #89 | Hy3 preview low | Tencent | 5.9 | 6.4 | 1/3 | 40.4s |
| #97 | Gemini 2.5 Flash none | 5.9 | 6.2 | 1/3 | 495ms |