Model Benchmark ArenaBlind Battles
WilBot AI Model Arena
Pit frontier neural models side-by-side in blind challenges. Test responses, vote on accuracy and reasoning, and track live community Elo rankings.
Global AI Model Arena Rankings
Ranked by Elo rating from community blind battles
| # | Model | Provider Engine | Arena Elo | Win Rate | Avg Speed |
|---|---|---|---|---|---|
| 🥇 | Gemini 3.6 Flash | Google DeepMind | 1342 | 78% | 420ms |
| 🥈 | Llama 3.3 70B Versatile | Groq LPU Engine | 1295 | 74% | 180ms |
| 🥉 | Qwen 2.5 72B Instruct | Alibaba / OpenRouter | 1260 | 71% | 650ms |
| #4 | Llama 3.1 8B Ultra-Fast | Cerebras CS-3 Core | 1180 | 65% | 95ms |