The spread between cheap and premium models
The distribution of token prices in the index, from the cheapest decile to the top percentile.
What the data shows
The median input price is $4.00 per million tokens, while the 90th percentile is $16.82. The top end of the market is far more expensive than the middle.
Output prices are consistently higher and have a wider spread. Long-form workloads should budget on output tokens, not input tokens alone.
How to read it
- Compare a model to the median, not the cheapest outlier.
- Estimate your input/output ratio.
- Check cached input pricing for repeated prompts.
- Leave room for retries and evaluation traffic.
Price distribution
| Percentile | Input price | Output price |
|---|---|---|
| 10th percentile | $0.65 | $2.78 |
| 25th percentile | $1.46 | $6.65 |
| 50th percentile | $4.00 | $18.20 |
| 75th percentile | $9.82 | $44.67 |
| 90th percentile | $16.82 | $73.51 |
| 99th percentile | $20.73 | $103.53 |
Methodology
This story uses the AI & LLM Index dataset: 600 models, 40 providers and 12 benchmarks. Prices are listed per million tokens. Scores are modelled estimates on a 0–100 scale and are not official benchmark results. Figures are rounded for readability.
Related
FAQ
What does the The spread between cheap and premium models data show?
The distribution of token prices in the index, from the cheapest decile to the top percentile.
Are the numbers official?
No. Prices and scores are compiled for this demo index. Benchmark values are modelled estimates, not official results. Verify figures with the provider or benchmark owner before making decisions.
How should I use this ranking?
Use it to form a shortlist. Check the specific model pages, compare price and context, and test the shortlist on your own examples before committing.