Kategoria ya AI BENCHY
Orodha ya Mahususi kwa domeni
Ona ni modeli gani za AI zinafanya vizuri zaidi katika Mahususi kwa domeni, zipi zinabaki thabiti, na pengo kubwa liko wapi.
Modeli zilizoonyeshwa
15
Wastani wa Alama ya Mahususi kwa domeni
4.8
Modeli bora
Gemini 3 Flash Preview 10.0| Nafasi | Modeli | Kampuni | Alama ya Mahususi kwa domeni | Alama | Majaribio sahihi | Muda wa majibu (wastani) |
|---|---|---|---|---|---|---|
| #152 | MiMo-V2-Flash none | Xiaomi | 5.3 | 4.6 | 1/3 | 564ms |
| #155 | Mercury 2 none | Inception | 5.3 | 4.5 | 1/3 | 534ms |
| #94 | GPT-5 Nano medium | OpenAI | 5.2 | 6.3 | 1/3 | 204.0s |
| #31 | DeepSeek V4 Flash high | DeepSeek | 4.1 | 7.7 | 0/3 | 100.3s |
| #45 | GPT-5.4 Mini medium | OpenAI | 4.1 | 7.5 | 0/3 | 65.3s |
| #66 | Qwen3.5-35B-A3B medium | Qwen | 4.1 | 7.1 | 0/3 | 88.3s |
| #71 | Step 3.7 Flash high | Stepfun | 4.1 | 7.0 | 0/3 | 149.6s |
| #107 | Laguna Xs.2 medium | Poolside | 4.1 | 5.8 | 0/3 | 11.1s |
| #18 | Qwen3.7 Plus medium | Qwen | 3.6 | 8.2 | 0/3 | 45.3s |
| #53 | Gemini 3.1 Flash Lite high | 3.6 | 7.3 | 0/3 | 139.9s | |
| #54 | GPT-5 Mini medium | OpenAI | 3.6 | 7.3 | 0/3 | 44.6s |
| #100 | Grok Build 0.1 none | X AI | 3.6 | 6.0 | 0/3 | 103.7s |
| #102 | Gemma 4 26B A4B none | 3.6 | 6.0 | 0/3 | 2.49s | |
| #110 | Seed-2.0-Lite none | Bytedance Seed | 3.6 | 5.8 | 0/3 | 1.33s |
| #141 | Nemotron 3 Super none | NVIDIA | 3.6 | 4.9 | 0/3 | 6.23s |