Kategoria ya AI BENCHY
Orodha ya Akili ya jumla
Ona ni modeli gani za AI zinafanya vizuri zaidi katika Akili ya jumla, zipi zinabaki thabiti, na pengo kubwa liko wapi. Panga kwa: Kipimo ↑.
| Nafasi | Modeli | Kampuni | Alama ya Akili ya jumla | Alama | Majaribio sahihi | Muda wa majibu (wastani) |
|---|---|---|---|---|---|---|
| #62 | Step 3.5 Flash medium | Stepfun | 5.5 | 7.2 | 0/1 | 22.4s |
| #71 | Step 3.7 Flash high | Stepfun | 5.5 | 7.0 | 0/1 | 4.17s |
| #84 | Grok 4.20 Multi Agent Beta medium | X AI | 5.8 | 6.6 | 0/1 | 6.40s |
| #17 | GLM 5 medium | Z.ai | 6.1 | 8.3 | 0/1 | 14.7s |
| #23 | GLM 5 Turbo medium | Z.ai | 6.1 | 8.0 | 0/1 | 10.1s |
| #30 | Qwen3.5-27B medium | Qwen | 6.1 | 7.8 | 0/1 | 101.4s |
| #31 | DeepSeek V4 Flash high | DeepSeek | 6.1 | 7.7 | 0/1 | 25.2s |
| #49 | Qwen3.5-Flash medium | Qwen | 6.1 | 7.4 | 0/1 | 40.1s |
| #77 | Claude Sonnet 4.6 none | Anthropic | 6.1 | 6.8 | 0/1 | 2.56s |
| #103 | DeepSeek V4 Pro high | DeepSeek | 6.1 | 6.0 | 0/1 | 25.1s |
| #116 | Hunter Alpha none | OpenRouter | 6.1 | 5.7 | 0/1 | 2.71s |
| #150 | Qwen3 Coder Next medium | Qwen | 6.3 | 4.6 | 0/1 | 1.39s |
| #76 | Kimi K2.5 medium | Moonshot AI | 6.5 | 6.8 | 0/1 | 69.7s |
| #78 | Qwen3.6 27B medium | Qwen | 6.5 | 6.8 | 0/1 | 39.5s |
| #117 | Qwen3.5-35B-A3B none | Qwen | 6.5 | 5.6 | 0/1 | 1.19s |