Free · On-Device · No API key

Claude Code on Apple Intelligence.

Route Claude Code through your Mac's Neural Engine. Free, private, and fast enough for everyday refactors and code review. No cloud round-trip, no per-token billing — Apple Intelligence runs entirely on-device.

Before you start

  • macOS 26+ on Apple Silicon (M1 or later) — Apple Intelligence is the dedicated Neural Engine path on these machines.
  • Claude Code installed (run <code class="font-mono text-xs text-white/70">claude --version</code> to verify).
  • ToolPiper installed (free DMG from modelpiper.com/download).

Setup

  1. 1

    Open ToolPiper and add an Apple Intelligence endpoint

    Launch ToolPiper. Open the menu-bar icon → Endpoints → "Add Endpoint" → pick Apple Intelligence. The Neural Engine backend lights up immediately — no model download, no warm-up. Name it whatever you want; the recipe assumes "Apple Intelligence (local)".

  2. 2

    Click "Configure for Claude Code"

    In ToolPiper, open Docs → Claude Code and press Configure for Claude Code. ToolPiper writes ~/.claude/settings.json with a fresh dev token, registers the MCP server, and adds your endpoint to Claude Code's /model picker. The whole step is reversible from the same panel.

  3. 3

    Open a terminal and run claude

    Open any terminal. The env vars are picked up from ~/.claude/settings.json on launch.

    shell
    claude
    # In Claude Code, type "/model" to confirm Apple Intelligence appears.
  4. 4

    Pick "Apple Intelligence (local)" in /model

    Claude Code's /model picker now lists your ToolPiper endpoints. Select Apple Intelligence — every subsequent prompt routes through the Neural Engine. Confirm with "What model are you?".

  5. 5

    Switch backends from inside Claude Code

    Type "switch to my local Qwen one" or "use my OpenAI endpoint" — ToolPiper's MCP endpoint_set tool reroutes the current shell. No file editing, no env var dance.

Why this recipe

Fully private

Apple Intelligence runs on the Neural Engine. Your prompts and code never leave the machine. Same hardware Apple uses for on-device summarization.

Zero ramp-up

No model download, no GPU warm-up. ToolPiper exposes Apple Intelligence the moment macOS does — first prompt is instant.

No bill, no quota

There's no token meter. Run Claude Code as much as you want — Apple Intelligence is included with macOS.

Frequently asked

How does Apple Intelligence compare to Claude Sonnet on hard tasks?

Apple Intelligence is built for fast on-device work — summaries, refactors, code review, mechanical edits. For multi-step reasoning or large-context refactors, switch to a stronger backend mid-session via <code class="font-mono text-xs text-white/70">"switch to my OpenAI endpoint"</code>. ToolPiper makes that one sentence; you keep your Claude Code session.

Will it work with Claude Code's tool use?

Yes. ToolPiper's Anthropic proxy translates Apple Intelligence's output into the Anthropic content-block shape Claude Code expects, including <code class="font-mono text-xs text-white/70">tool_use</code> blocks. Bash, Edit, Read, Write all flow through.

Do I need an Anthropic API key?

No. ToolPiper writes its own dev token to Claude Code's settings. The token validates locally against ToolPiper's token store — never reaches Anthropic.

How do I undo this?

In ToolPiper → Docs → Claude Code, click <em>Disconnect from Claude Code</em>. ToolPiper revokes the auto-token, removes its keys from <code class="font-mono text-xs text-white/70">~/.claude/settings.json</code>, and unregisters the MCP server. Your other settings stay untouched.

Ready to try it?

ToolPiper is a free download. Configure once and Claude Code routes through your Mac.

Other Claude Code recipes