- The comparator places two to four AI models side by side in one table.
- It compares prices, context window, capabilities, and the Artificial Analysis index.
- The best value in each row is highlighted, and your selection is saved in the link.
- Data comes from OpenRouter and updates daily, but testing on your own use cases remains essential.
Price, context window, capabilities: this comparator puts 2 to 4 AI models side-by-side in a single table, so you can choose without opening ten pages of documentation.
How to use it
- Search for a model by its name or publisher and add it to the comparison (up to four).
- Read the table: the best value in each row is highlighted.
- Copy the page address: your selection is saved in the link, handy for sharing a comparison.
Principles behind the calculations
- The rows cover the publisher, release date, input, output, and cache read pricing, context window, maximum output, accepted input and output types, tool calling, reasoning, structured outputs, availability of weights on Hugging Face, and the Artificial Analysis index.
- A ‘typical usage’ row compares the cost of one million input tokens and two hundred thousand output tokens, for a quick read on price-performance.
- For prices, the lowest is highlighted; for context, maximum output, and the index, the highest. If all models are equal, nothing is highlighted.
Limitations to be aware of
Data comes from OpenRouter and is updated daily. A performance index does not replace testing on your own use cases.
