Skip to content

DirectoryTools

MiniMax M2.1 logo

MiniMax M2.1

MiniMax's open-weight coding model tuned for Rust, Java, Go, C++, Kotlin and mobile app work.

Visit MiniMax M2.1

Coding & WebFreeWe checked the prices on platform.minimax.io on 7 Oct 2026.

MiniMax M2.1 homepage screenshot

What is MiniMax M2.1?

MiniMax's open-weight coding model tuned for Rust, Java, Go, C++, Kotlin and mobile app work.

Coding & WebModels & Open WeightsModelOpen weightsCoding

Who is MiniMax M2.1 best for?

MiniMax M2.1 is best for coders who want an open model for Android and iOS work.

  • Multi-language codebases

    MiniMax trained it beyond Python, with stronger Rust, Java, Go, C++, Kotlin, Objective-C, TypeScript and JavaScript.

  • Native mobile and web apps

    It improves on native Android and iOS development and on the look of the web and app interfaces it builds.

  • Coding agents you already use

    MiniMax reports steady results in Claude Code, Droid, Cline, Kilo Code, Roo Code and BlackBox, with support for Skill.md, agent rule files and slash commands.

  • Self-hosted deployments

    The weights are on Hugging Face, with deployment guides for SGLang, vLLM and Transformers and a tool-calling guide.

How much does MiniMax M2.1 cost?

MiniMax M2.1's weights are free to download under a modified MIT license, and the MiniMax API charges $0.30 per million input tokens and $1.20 per million output tokens.

Open weights$0Modified MIT license
  • Download from Hugging Face
  • Serve with SGLang, vLLM or Transformers
  • Commercial use must credit MiniMax M2.1 in the interface
MiniMax-M2.1 API$0.30 / $1.20Per million input / output tokens
  • Cached reads $0.03 per million tokens
  • Cache writes $0.375 per million tokens
MiniMax-M2.1-highspeed API$0.60 / $2.40Per million input / output tokens
  • Faster version of the same model
  • Same cache prices as the standard model

MiniMax now lists M2.1 under its legacy models on the pay-as-you-go price page; newer M2.5 and M2.7 models cost the same per token. Full pricing on platform.minimax.io. We checked these prices on .

How do readers rate MiniMax M2.1?

No one has rated MiniMax M2.1 yet. Be the first.

No ratings yetThe score and its stars show once 3 readers have rated MiniMax M2.1.
Your rating

Sign in to rate. One rating per account.

Your rating of MiniMax M2.1

What are the best MiniMax M2.1 alternatives?

Three tools on Recaply do MiniMax M2.1's job. GLM-4.7 and Qwen3-Coder-Next are open source, so you can run them on your own server.

Questions about MiniMax M2.1

What is MiniMax M2.1?

MiniMax's open-weight coding model tuned for Rust, Java, Go, C++, Kotlin and mobile app work.

Who is MiniMax M2.1 for?

MiniMax M2.1 is best for coders who want an open model for Android and iOS work.

Is MiniMax M2.1 free?

Yes. MiniMax M2.1 has a free Open weights plan: download from Hugging Face.

How much does MiniMax M2.1 cost?

MiniMax M2.1's paid plans are MiniMax-M2.1 API at $0.30 / $1.20 per million input / output tokens and MiniMax-M2.1-highspeed API at $0.60 / $2.40 per million input / output tokens.

Is there an open source alternative to MiniMax M2.1?

Yes. GLM-4.7 and Qwen3-Coder-Next are open source tools that do the same job, so you can run your own.

What are the best MiniMax M2.1 alternatives?

The closest alternatives on Recaply are GLM-4.7, Qwen3-Coder-Next and Kimi K2.7 Code.

Where can I try MiniMax M2.1?

On its official website, minimax.io.

Can I use MiniMax M2.1 commercially?

Yes. The weights use MIT terms with one change: if a commercial product or service uses M2.1 or a model derived from it, it must prominently show the name MiniMax M2.1 in its user interface.