AI BENCHY Category
Anti-AI Tricks Ranking
See which AI models perform best on Anti-AI Tricks, which ones stay reliable, and where the biggest gaps appear. Sort by: Tests Correct ↑.
| Rank | Model | Company | Anti-AI Tricks Score | Score | Tests Correct | Response Time (avg) |
|---|---|---|---|---|---|---|
| #149 | Nemotron 3 Nano Omni 30b A3b Reasoning medium | NVIDIA | 6.4 | 4.6 | 2/4 | 1.20s |
| #8 | Claude Opus 4.7 none | Anthropic | 8.3 | 8.9 | 3/4 | 2.12s |
| #11 | Claude Opus 4.7 medium | Anthropic | 8.3 | 8.7 | 3/4 | 1.85s |
| #13 | Grok 4.20 Beta medium | X AI | 8.7 | 8.5 | 3/4 | 3.16s |
| #15 | GPT-5.3-Codex medium | OpenAI | 8.7 | 8.4 | 3/4 | 4.16s |
| #19 | Seed-2.0-Lite medium | Bytedance Seed | 8.3 | 8.2 | 3/4 | 18.0s |
| #21 | GPT-5.4 medium | OpenAI | 8.3 | 8.0 | 3/4 | 4.11s |
| #22 | Step 3.7 Flash medium | Stepfun | 8.7 | 8.0 | 3/4 | 9.65s |
| #24 | GPT-5.2 Chat none | OpenAI | 8.7 | 7.9 | 3/4 | 3.40s |
| #25 | Qwen3.5 Plus 2026-02-15 medium | Qwen | 8.2 | 7.9 | 3/4 | 45.8s |
| #28 | Gemini 2.5 Flash medium | 8.4 | 7.8 | 3/4 | 6.30s | |
| #30 | Qwen3.5-27B medium | Qwen | 8.7 | 7.8 | 3/4 | 19.8s |
| #31 | DeepSeek V4 Flash high | DeepSeek | 8.3 | 7.7 | 3/4 | 28.5s |
| #40 | Gemini 3.1 Flash Lite Preview medium | 9.1 | 7.5 | 3/4 | 2.33s | |
| #44 | Gemini 3.1 Flash Lite medium | 9.1 | 7.5 | 3/4 | 2.39s |