Model rankings
Published benchmark scores for the models you can call through TokenCenter, next to what they cost. Use them to find the best model for your budget.
Loading…
| # | Model | Intelligence | Coding | Agentic | Input / 1M | Output / 1M |
|---|
Scores are Artificial Analysis indexes (higher is better) as published in OpenRouter’s model data, fetched Oct 15, 2026. TokenCenter does not run these benchmarks. Models without a published score aren’t listed; batch variants share their base model’s score and are hidden. Prices are US dollars per 1M tokens at provider cost.
| Arena rank | Model | Elo | Win rate | Input / 1M | Output / 1M |
|---|
Design Arena results (head-to-head human votes on generated work; higher Elo and win rate are better) as published in OpenRouter’s model data, fetched Oct 15, 2026. Arena rank is Design Arena’s own position across all models it tests, so numbers can skip models that TokenCenter doesn’t offer.