Skip to content

How to Run Xiaomi MiMo V2.5 Pro With Claude Code

Run Xiaomi MiMo V2.5 Pro with Claude Code for cheap, big-context coding. API key, endpoint setup, model choice, current V2.5 API pricing, free trials, and the fixes.

MGMCSA Guru Team July 15, 2026 3 min read
Claude Code on Windows running Xiaomi MiMo V2.5 Pro

Xiaomi’s MiMo V2.5 Pro is a strong budget pick for one specific reason: a very large context window (around 1M tokens) paired with low, cache-aware API pricing and — notably — no surcharge for using that big context. For agentic coding on large repositories, being able to feed a lot of code without a long-context penalty is genuinely useful, and it pairs well with Claude Code.

This covers connecting MiMo to Claude Code, the model lineup, and the pricing. For routing options generally, see use cheap AI models with Claude Code.

Step 1: Get a MiMo API key

Sign up on Xiaomi’s MiMo platform and create an API key. Watch for free trial windows — MiMo has been offered with free access through various coding tools at launch.

Step 2: Connect to Claude Code

Two paths, depending on what MiMo exposes:

Anthropic-compatible endpoint (if available — check the docs): point Claude Code straight at it:

export ANTHROPIC_BASE_URL="https://api.xiaomimimo.com/anthropic"
export ANTHROPIC_AUTH_TOKEN="your-mimo-key"
export ANTHROPIC_MODEL="mimo-v2.5-pro"
claude

Otherwise, via Claude Code Router using MiMo’s OpenAI-compatible endpoint:

{
  "Providers": [
    {
      "name": "mimo",
      "api_base_url": "https://api.xiaomimimo.com/v1/chat/completions",
      "api_key": "$MIMO_API_KEY",
      "models": ["mimo-v2.5-pro"]
    }
  ],
  "Router": { "default": "mimo,mimo-v2.5-pro" }
}

Step 3: Lean on the big context

MiMo’s ~1M token context is the reason to choose it. With no long-context surcharge, you can load large portions of a codebase affordably. In a router setup, set a high longContextThreshold so big-repo requests stay on MiMo.

MiMo models (confirm on Xiaomi)

mimo-v2.5-pro Flagship coding model; ~1M context
mimo-v2.5 Standard tier
mimo-v2-flash Faster, lighter variant

Pricing

Xiaomi’s official V2.5 price update lists MiMo-V2.5-Pro at $0.0036 cache-hit input / $0.435 cache-miss input / $0.87 output per 1M tokens. Standard MiMo-V2.5 is cheaper at $0.0028 / $0.14 / $0.28 per 1M tokens. Xiaomi says legacy MiMo-V2-Pro / Omni names auto-route to V2.5 pricing from June 1, 2026 and are deprecated on June 30, 2026. Full detail: Xiaomi MiMo pricing and free tiers.

Verify

claude
# then: "load these files and explain how the module fits together"

If it handles the large context cleanly, MiMo is serving Claude Code.

Troubleshooting

  • Auth errors — key not set in the current shell.
  • Format errors — confirm the Anthropic vs OpenAI endpoint matches your setup path.
  • “Model not found” — verify the model ID on Xiaomi’s platform.
  • Config not applied (router) — run ccr restart.

MiMo + Claude Code checklist

  • MiMo API key created (check for free trials)
  • Claude Code installed
  • Connected via Anthropic endpoint or Claude Code Router
  • Model set to mimo-v2.5-pro (verify ID)
  • Big context exploited for large repos

Wrapping up

Xiaomi MiMo V2.5 Pro with Claude Code is the setup for big-repo coding on a budget: connect via an Anthropic endpoint or Claude Code Router, set the model, and exploit the ~1M context window — which, with no long-context surcharge and cache-aware V2.5 pricing, stays cheap to fill. Verify the current endpoint and rates on Xiaomi’s platform.

To use MiMo with Codex, see MiMo V2.5 + Codex CLI on Windows; for the cost detail, see Xiaomi MiMo pricing and free tiers.

Frequently asked questions

What is Xiaomi MiMo V2.5 Pro?

It's Xiaomi's flagship MiMo coding model — a large mixture-of-experts model with a very large (around 1M token) context window, aimed at agentic coding. It's positioned as cheap and strong for its tier.

How do I connect MiMo to Claude Code?

MiMo exposes an API on Xiaomi's platform. If it offers an Anthropic-compatible endpoint, point Claude Code at it with environment variables; otherwise use Claude Code Router with MiMo's OpenAI-compatible endpoint. Check Xiaomi's docs for the current endpoints.

How much does MiMo cost?

As of Xiaomi's May 27, 2026 V2.5 price update, MiMo-V2.5-Pro is $0.0036 per 1M cache-hit input tokens, $0.435 per 1M cache-miss input tokens, and $0.87 per 1M output tokens. Standard MiMo-V2.5 is cheaper at $0.0028 / $0.14 / $0.28 per 1M tokens.

Are there free trials?

MiMo has been offered with free trial access through various coding tools (such as OpenCode, Cline, and others) at launch windows. Availability changes, so check Xiaomi's platform and the tool integrations for current free access.

Why pick MiMo for coding?

Its very large context window suits big repositories and long sessions, and the flat low pricing with no long-context penalty makes feeding lots of code affordable — a combination not every cheap model offers.

Sources & further reading

Official vendor documentation referenced while writing this guide.

MG

MCSA Guru Team

IT & Systems Administration

We are working IT pros and system administrators who spend our days in Windows Server, Microsoft 365, and the wider Microsoft stack. MCSA Guru is where we write down the fixes and walkthroughs we wish we had found the first time.

MCSA Guru provides independent, educational IT guidance. Microsoft, Windows, Windows Server, Microsoft 365, Exchange, and Microsoft Teams are trademarks of Microsoft Corporation; Docker is a trademark of Docker, Inc. MCSA Guru is not affiliated with or endorsed by Microsoft or Docker. Always test changes in a safe environment before applying them in production.

Related guides

Fixing something right now?

Jump straight into the guide library or search for the exact error or task you are dealing with.