Skip to main content

Using MiMo With Zoo Code

Zoo Code supports accessing models through the Xiaomi MiMo API, including mimo-v2.5-pro and mimo-v2.5. MiMo models are designed for deep reasoning, tool calling, and structured output with support for up to 1 million tokens of context.

Website: https://platform.xiaomimimo.com/

Getting an API Key​

  1. Sign Up/Sign In: Go to the MiMo Platform. Create an account or sign in.
  2. Navigate to API Keys: Find your API keys in the platform dashboard.
  3. Create a Key: Click "Create new API key". Give your key a descriptive name (e.g., "Zoo Code").
  4. Copy the Key: Important: Copy the API key immediately. You will not be able to see it again. Store it securely.

Available Models​

Zoo Code supports the following MiMo models:

ModelContext WindowMax OutputImagesDescription
mimo-v2.5-pro (default)1M tokens131K tokensNoFlagship reasoning model with deep thinking and tool calling
mimo-v2.51M tokens131K tokensYesFull-modal model (text, image, audio, video input) with deep thinking

For the complete, up-to-date model list and capabilities, see MiMo's API documentation.

Configuration in Zoo Code​

  1. Open Zoo Code Settings: Click the gear icon () in the Zoo Code panel.
  2. Select Provider: Choose "Xiaomi MiMo" from the "API Provider" dropdown.
  3. Enter API Key: Paste your MiMo API key into the "MiMo API Key" field.
  4. Select Model: Choose your desired model from the "Model" dropdown.
  5. (Optional) Base URL: If you need to use a custom base URL for the MiMo API, enter it in the "Base URL" field. The default is https://token-plan-sgp.xiaomimimo.com/v1. Other regional endpoints are also available (e.g., https://token-plan-cn.xiaomimimo.com/v1 for China, https://token-plan-ams.xiaomimimo.com/v1 for Europe).

Tips and Notes​

  • Pricing: MiMo offers competitive pricing. See the MiMo pricing page for current rates. Note that long-context usage (above 256K tokens) is charged at 2x the standard rate.
  • Thinking Mode: MiMo models support deep thinking mode which is enabled by default. The model reasons through complex problems step by step, improving accuracy on coding and reasoning tasks.
  • Tool Calling: MiMo models support tool calling for agentic workflows. Reasoning content is preserved across multi-turn conversations with tool calls for consistent results.
  • Cache Pricing: Cache reads are available at a discounted rate. Cache writes are currently free for a limited time.