Xiaomi MiMo API Pricing Calculator & Cost Guide

Xiaomi MiMo is billed per token, with separate input (prompt) and output (completion) rates that vary by model. MiMo is positioned as a budget open-weight family, so its rates sit at the low end of the market. Use the calculator above for live per-model rates and a monthly estimate for your workload.

Each MiMo model has its own input and output rate per million tokens, and output usually costs more than input. The two models in the catalog, MiMo v2.5 and MiMo v2.5 Pro, are priced as budget options. The table above shows the current per-million-token rates for each, pulled live so you always see what you would actually pay.

Context is one of MiMo's headline features. The models support a very long context window, up to roughly 1 million tokens, which lets you feed in large documents, long codebases, or extended chat history in a single call without chunking. That makes MiMo a fit for long-context tasks where cheaper short-context models fall short.

MiMo ships with open weights, so you can download and self-host the models at no per-token cost if you have the hardware. For hosted API access you pay per token, and some providers offer rate-limited free access for testing through OpenRouter. Open weights plus low hosted pricing is the core of MiMo's value.

This page is about Xiaomi MiMo, the open large language model family from Xiaomi, not the Mimo app that teaches people to code. As a budget model, Xiaomi MiMo competes on price and context length rather than topping the quality leaderboard, so compare it against other low-cost long-context models like DeepSeek or Qwen on your specific task rather than on headline price alone.