MiniMax API Pricing Calculator & Cost Guide

MiniMax charges per token for its text models, with separate input (prompt) and output (completion) rates that vary by model. The MiniMax-01 family and MiniMax-M2 each carry their own rates, and long prompts cost more because you pay for every input token. Use the calculator above for live per-model rates and a monthly estimate for your workload.

MiniMax-M2 is billed per million tokens, with input and output priced separately like the rest of the MiniMax text lineup. It targets agentic and long-context work, so cost depends heavily on how much context you feed it per call. Check the table above for the current M2 input and output rates and plug your token counts into the calculator.

MiniMax positions its text models as low-cost options for agentic and long-context use, and they generally sit well below Western frontier flagships on price. Whether it is the cheapest for your task depends on how many input tokens your long-context prompts consume, since large context windows can drive real cost even at a low per-token rate. Compare on your actual token profile using the calculator.

No. This page covers only MiniMax's text and LLM endpoints. MiniMax also sells audio, text-to-speech, and Hailuo video generation, and those media products are billed on their own separate pricing, usually per second, per character, or per generation rather than per text token. Budget for them separately from the text rates shown here.

The live rates in the table come from the OpenRouter model catalog, which passes through provider pricing at or near direct cost, and refresh daily. Always confirm against MiniMax's own platform pricing before committing to production budgets, since providers update rates and add models regularly.