context window

Context window comparison

How much can each model read at once? Sort 300+ models by context window to find the ones that fit a whole book, a large codebase or a long chat history, alongside what they cost.

#ModelProviderContext$/M in$/M out
Loading models...

Live data via OpenRouter. Context is the maximum tokens (input + output) a model can handle per request. A bigger window costs more per call when you fill it.