re:cinq Logore:cinq

Running MiniMax M2.5 Locally with Claude Code

Running MiniMax M2.5 Locally with Claude Code

TL;DR: Once you have MiniMax M2.5 running locally (see previous post), here's how to connect Claude Code to it.

If you've got MiniMax M2.5 running via the setup guide, connecting Claude Code is straightforward.

Basic Configuration

Edit your ~/.claude.settings.json:

{
  "env": {
    "ANTHROPIC_BASE_URL": "http://<your-server>:8080",
    "ANTHROPIC_AUTH_TOKEN": "any-placeholder-value",
    "ANTHROPIC_MODEL": "MiniMax-M2.5",
    "ANTHROPIC_SMALL_FAST_MODEL": "MiniMax-M2.5",
    "ANTHROPIC_DEFAULT_SONNET_MODEL": "MiniMax-M2.5",
    "ANTHROPIC_DEFAULT_OPUS_MODEL": "MiniMax-M2.5",
    "ANTHROPIC_DEFAULT_HAIKU_MODEL": "MiniMax-M2.5",
    "CLAUDE_CODE_SUBAGENT_MODEL": "MiniMax-M2.5",
    "API_TIMEOUT_MS": "3000000",
    "CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC": "1"
  }
}

Replace <your-server> with your inference server's address.

With Agent Teams and Hooks

If you use agent teams, skills, or hooks:

{
  "env": {
    "CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS": "1",
    "ANTHROPIC_BASE_URL": "http://<your-server>:8080",
    "ANTHROPIC_AUTH_TOKEN": "local",
    "ANTHROPIC_MODEL": "MiniMax-M2.5",
    "ANTHROPIC_SMALL_FAST_MODEL": "MiniMax-M2.5",
    "ANTHROPIC_DEFAULT_SONNET_MODEL": "MiniMax-M2.5",
    "ANTHROPIC_DEFAULT_OPUS_MODEL": "MiniMax-M2.5",
    "ANTHROPIC_DEFAULT_HAIKU_MODEL": "MiniMax-M2.5",
    "CLAUDE_CODE_SUBAGENT_MODEL": "MiniMax-M2.5",
    "API_TIMEOUT_MS": "3000000",
    "CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC": "1"
  },
  "hooks": {
    "UserPromptSubmit": [
      {
        "hooks": [
          {
            "type": "command",
            "command": "$HOME/.claude/hooks/skill-activation-prompt.sh"
          }
        ]
      }
    ],
    "PreToolUse": [
      {
        "matcher": "Edit|MultiEdit|Write",
        "hooks": [
          {
            "type": "command",
            "command": "$HOME/.claude/hooks/skill-verification-guard.sh"
          }
        ]
      }
    ],
    "PostToolUse": [
      {
        "matcher": "Edit|MultiEdit|Write",
        "hooks": [
          {
            "type": "command",
            "command": "$HOME/.claude/hooks/post-tool-use-tracker.sh"
          }
        ]
      }
    ]
  },
  "skipDangerousModePermissionPrompt": true
}

Next Steps

  1. Start Claude Code with your configuration
  2. If prompted, log out and log back in to clear any warnings
  3. You're running entirely on local hardware

This gives you a local, private inference endpoint that works with Claude Code - useful for analyzing sensitive codebases without cloud dependencies.

Resources

Related Articles

  • Running MiniMax M2.5 Locally on NVIDIA DGX Spark

    Michael Mueller10 min read

    How I got a 230B parameter open model running on desktop hardware using NVIDIA D…

  • Meet Reachy Mini: Building an AI-Powered Conference Badge Reader

    Michael Mueller4 min read

    How I built a fun conference booth experience combining an open-source robot, vi…

  • Multi-Agent Orchestration: BMAD, Claude Flow, and Gas Town

    Michael Mueller10 min read

    A deep dive into three multi-agent AI orchestration frameworks: BMAD for structu…

See all articles

You Might Also Like

A Pattern Language for Transformation

Get a shared vocabulary of proven Transformation Patterns, common Anti-Patterns, and Paradigm Patterns to have more effective, data-driven conversations about your strategy and architecture.

Explore the Patterns →

Free AI Assessment

For a personalized starting point, take our free online assessment. Your results will give you a detailed report on your current maturity and suggest the most relevant patterns to focus on first.

Take the Free Assessment →

Waves of Innovation

Every Tuesday, we deliver one short, powerful read on AI Native to help you lead better, adapt faster, and build smarter—based on decades of experience helping teams transform for real.

Get the Report →