Skip to content
OpenKey

LLM API price comparison

All 338 models in one live table. Prices are per 1M tokens and already include the flat 3% fee — the provider price behind every number is printed on each model page and on /pricing. Set your monthly workload to rank models by what they would actually cost you.

338 of 338 models · workload 10M in + 2M out · flat 3% fee included

Workload cost = input price × input millions + output price × output millions. Provider prices sourced live from the catalog; variable-priced routers are excluded from cost ranking because their price depends on the routed model.

Keep comparing

Every price here is one key away.

Switch between any of these 338 models by changing one string in an OpenAI-compatible request.