Baltor Get started

Models

MiMo-V2.6-Flash

By Xiaomi. Unknown parameters, a context window of 1,048,576 tokens, licence Unknown.

Facts

Released
2026-09-22 the release date models.dev records models.dev, read
Licence
Unknown
Open weights
Yes models.dev records whether the weights are open models.dev, read
Parameters
Unknown
Active parameters
Unknown
Knowledge cutoff
Unknown
Context window
  • 1,048,576 tokens the context limit models.dev records for the maker's own API models.dev, read
  • 1,048,576 tokens the input limit LiteLLM records for the maker's own API LiteLLM, read
Longest output
  • 131,072 tokens the output limit models.dev records for the maker's own API models.dev, read
  • 131,072 tokens the output limit LiteLLM records for the maker's own API LiteLLM, read
Tool calling
  • Yes models.dev records tool_call as true models.dev, read
  • Yes LiteLLM records supports_function_calling as true for the maker's own API LiteLLM, read
Structured output
  • Yes LiteLLM records supports_response_schema as true for the maker's own API LiteLLM, read
Reasoning controls
  • Yes models.dev records reasoning as true models.dev, read
  • Yes LiteLLM records supports_reasoning as true for the maker's own API LiteLLM, read
Inputs and outputs
text, image, audio, video in, text out models.dev, read
Good for
  • Vision: models.dev lists images among its inputs models.dev, read
  • Reasoning: models.dev records it as a reasoning model models.dev, read

Prices

Each provider's own price in US dollars per million tokens, as an openly licensed source lists it
ProviderInputOutputContextAs ofSource
Xiaomi0.140 USD0.280 USD1,048,576models.dev, read

Run it on your own hardware

No source lists weight files or a parameter count for this model, so its memory needs are Unknown.

Use it from your harness

Set up for Xiaomi with the model mimo-v2.6-flash. Each endpoint page has the same setup for its own address.

OpenCode

Put this in opencode.json in your project folder:

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "xiaomi": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "Xiaomi",
      "options": {
        "baseURL": "https://api.xiaomimimo.com/v1",
        "apiKey": "{env:XIAOMI_API_KEY}"
      },
      "models": {
        "mimo-v2.6-flash": {
          "name": "MiMo-V2.6-Flash",
          "limit": {
            "context": 1048576,
            "output": 131072
          }
        }
      }
    }
  }
}
  • OpenCode reads any OpenAI-compatible address through the @ai-sdk/openai-compatible package, and an address that speaks the Responses API through @ai-sdk/openai.

From OpenCode documentation, read .

Pi

Put this in ~/.pi/agent/models.json:

{
  "providers": {
    "xiaomi": {
      "baseUrl": "https://api.xiaomimimo.com/v1",
      "api": "openai-completions",
      "apiKey": "$XIAOMI_API_KEY",
      "models": [
        {
          "id": "mimo-v2.6-flash"
        }
      ]
    }
  }
}
  • The apiKey field can name an environment variable as $NAME.

From Pi documentation, read .

Codex

Codex speaks only the Responses API, and Xiaomi documents no Responses address. A gateway that offers one can sit in between.

  • Codex speaks the Responses API only: responses is the one supported wire API of a custom provider. Ollama and LM Studio are built in and start with --oss.

From Codex documentation, read .

Claude Code

Claude Code sends Anthropic Messages requests, and Xiaomi documents no such address. A gateway that translates to that API can sit in between.

  • Claude Code sends Anthropic Messages requests to ANTHROPIC_BASE_URL. Anthropic says it does not support routing Claude Code to models other than Claude through any gateway, so some features may not work with another model.

From Claude Code documentation, read .

Published results

These are results other people published. Baltor did not run them and does not rank models by them.

The arena score is the overall text score with style control from LMArena's latest leaderboard, rounded to a whole number.

Two more publishers are linked, never copied, because their terms do not allow it:

  • Artificial Analysis publishes its own intelligence, speed and price measurements.
  • OpenRouter lists current prices and uptime for each provider it routes to.