AI BENCHY Category
Coding Ranking
See which AI models perform best on Coding, which ones stay reliable, and where the biggest gaps appear. Sort by: Tests Correct ↓.
| Rank | Model | Company | Coding Score | Score | Tests Correct | Response Time (avg) |
|---|---|---|---|---|---|---|
| #17 | Qwen3.5-122B-A10B medium | Qwen | 4.7 | 8.1 | 0/1 | 71.0s |
| #18 | Qwen3.6 Plus medium | Qwen | 3.0 | 8.1 | 0/1 | 0ms |
| #22 | Gemma 4 26B A4B medium | 2.8 | 8.0 | 0/1 | 147.5s | |
| #25 | DeepSeek V3.2 medium | DeepSeek | 4.7 | 8.0 | 0/1 | 180.9s |
| #30 | Qwen3.5-Flash medium | Qwen | 4.7 | 7.8 | 0/1 | 45.7s |
| #31 | GLM 5.1 medium | Z.ai | 4.7 | 7.8 | 0/1 | 118.5s |
| #32 | MiMo-V2-Omni medium | Xiaomi | 4.0 | 7.7 | 0/1 | 68.5s |
| #38 | MiMo-V2-Flash medium | Xiaomi | 4.7 | 7.5 | 0/1 | 13.0s |
| #43 | Kimi K2.5 medium | Moonshot AI | 4.7 | 7.0 | 0/1 | 150.8s |
| #44 | Grok 4.20 medium | X AI | 4.3 | 7.0 | 0/1 | 24.3s |
| #46 | Qwen3.5 Plus 2026-02-15 none | Qwen | 6.3 | 6.8 | 0/1 | 3.63s |
| #47 | Hunter Alpha medium | OpenRouter | 3.0 | 6.7 | 0/1 | 0ms |
| #48 | Nemotron 3 Super medium | NVIDIA | 3.0 | 6.7 | 0/1 | 0ms |
| #49 | Grok 4.1 Fast medium | X AI | 2.3 | 6.7 | 0/1 | 23.6s |
| #50 | GLM 5 none | Z.ai | 5.6 | 6.6 | 0/1 | 8.84s |