Get started
Text model

MiMo by Xiaomi — a million tokens for very little

MiMo models by Xiaomi in GPTunneL: a 1M token context, reasoning and image understanding at one of the lowest prices. Pay per token, no VPN.

What the MiMo family is good at

The traits Xiaomi keeps from generation to generation — you can count on them whichever model you pick.

Priced like the cheapest models

The main reason people pick this family: a million tokens of context and reasoning at a price comparable to compact models.

A very long context

The current models hold a million tokens — a folder of documents or an entire project, with no slicing into chunks.

Reasoning mode

The model spends more time on a hard task and reaches the answer via intermediate steps — it helps with maths, logic and reading code.

Reads images

The family is multimodal — an attached screenshot, diagram or photo is read alongside the text of the request.

Which model to pick

Two current models in the family — they differ in depth of analysis and price.

MiMo-V2.5The default pickMiMo-V2.5-ProHard tasks
Best forEveryday text, reading documents, imagesHard reasoning, large projects, agentic runs
Context1M tokens1M tokens
Reads imagesYesNo
Price per tokenLowestLow

Specs come from the platform catalogue. Exact per-model prices are on the Pricing

Every MiMo model

Release dates follow Xiaomi's official announcements.

April 2026

MiMo-V2.5MiMo-V2.5-Pro

The current generation: context grew to a million tokens. A larger Pro variant sits alongside it — stronger on hard reasoning, but it does not read images.

March 2026

MiMo-V2-OmniMiMo-V2-Pro

The generation where the family went multimodal: Omni read text, images, video and audio in one architecture. Xiaomi retired these models in the summer of 2026, replacing them with version 2.5.

2025

MiMo

The first generation, the one Xiaomi entered the open language model market with.

How billing works

There is no subscription: you top up one balance and spend it on any model on the platform.

Pay per token

You are charged for exactly what the request and the answer used. Not using it costs nothing: no limits, no monthly fee.

One balance for every model

MiMo, ChatGPT, Claude, image and video generation — all out of the same wallet. No separate subscription per service.

Top up the way you prefer

International cards, Apple Pay, Google Pay or crypto. The minimum top-up is $5.

See pricing

How to start with MiMo

Sign in to GPTunneL
One account for every model on the platform.
What is wrong with this approval flow and where will it loop in practice
flow.pngMiMo-V2.5
Flow review

The rejection branch returns the request to the same step it came from — a second rejection closes the loop.

There is no exit condition on iteration count, and no deadline after which the request escalates.

Parallel approvals are drawn in sequence, so in practice the turnaround will be twice the planned figure.

Sign in whichever way suits you.

The model switches right inside the input — you can change it mid-conversation.

The answer arrives in the chat and the context is kept until the conversation ends.

Try MiMo in GPTunneL

Signing up takes a minute, and $5 on the balance is enough to see whether the model fits your task.

Frequently asked questions

No. Requests go through GPTunneL's infrastructure, so the models open like any ordinary website — no VPN and no proxy.

MiMo is Xiaomi's family of language models — from the same company that makes phones and home appliances. On the model market it stands out mostly on the ratio of capability to price: a million tokens of context, reasoning and image understanding cost roughly what compact models do.

The current MiMo-V2.5 generation shipped in two variants, regular and Pro. The regular one reads images alongside text; Pro is stronger on hard reasoning. The previous V2 generation was retired in the summer of 2026 and replaced by version 2.5. Xiaomi publishes the weights openly, so MiMo can also be run on your own hardware.

MiMo models are available in GPTunneL — worldwide, without a VPN, with one balance and billing for the tokens you actually spend.