[BUG] v2.0.76 on Apple Silicon: AVX warning + spawns 40+ zombie node processes causing OOM

Open 💬 7 comments Opened Jan 1, 2026 by yonatangross

Preflight Checklist

  • [x] I have searched existing issues and this hasn't been reported yet
  • [x] This is a single bug report (please file separate reports for different bugs)
  • [x] I am using the latest version of Claude Code

What's Wrong?

Bug Description

Claude Code v2.0.76 on Apple Silicon (M4 Max) shows an x64 Bun warning and spawns dozens of zombie node processes on startup, causing memory exhaustion.

Environment

  • OS: macOS (Apple Silicon M4 Max, 256GB RAM)
  • Claude Code Version: 2.0.76
  • Installation Method: curl -fsSL https://claude.ai/install.sh | sh

Symptoms

1. Wrong architecture warning on every launch:

warn: CPU lacks AVX support, strange crashes may occur. Reinstall Bun or use *-baseline build:
  https://github.com/oven-sh/bun/releases/download/bun-v1.3.5/bun-darwin-x64-baseline.zip

This suggests the bundled Bun runtime is x64, not arm64.

2. Node process explosion:

Simply launching claude --dangerously-skip-permissions spawns 40+ node processes (~70-77MB each), totaling 3-4GB of RAM. These processes persist even after /exit.

Steps to Reproduce

  1. Install Claude Code on Apple Silicon Mac
  2. Run claude --dangerously-skip-permissions
  3. Observe AVX warning
  4. Open Activity Monitor → filter by "node"
  5. See 40+ node processes spawned immediately

Expected Behavior

  • No AVX warning (should use arm64 Bun)
  • Reasonable number of node processes that clean up on exit

Screenshots

[Activity Monitor showing 40+ node processes at ~70MB each]

Workaround

pkill -9 node to kill zombie processes, but they respawn on next launch## Bug Description

Claude Code v2.0.76 on Apple Silicon (M4 Max) shows an x64 Bun warning and spawns dozens of zombie node processes on startup, causing memory exhaustion.

Environment

  • OS: macOS (Apple Silicon M4 Max, 256GB RAM)
  • Claude Code Version: 2.0.76
  • Installation Method: curl -fsSL https://claude.ai/install.sh | sh

Symptoms

1. Wrong architecture warning on every launch:

warn: CPU lacks AVX support, strange crashes may occur. Reinstall Bun or use *-baseline build:
  https://github.com/oven-sh/bun/releases/download/bun-v1.3.5/bun-darwin-x64-baseline.zip

This suggests the bundled Bun runtime is x64, not arm64.

2. Node process explosion:

Simply launching claude --dangerously-skip-permissions spawns 40+ node processes (~70-77MB each), totaling 3-4GB of RAM. These processes persist even after /exit.

Steps to Reproduce

  1. Install Claude Code on Apple Silicon Mac
  2. Run claude --dangerously-skip-permissions
  3. Observe AVX warning
  4. Open Activity Monitor → filter by "node"
  5. See 40+ node processes spawned immediately

Expected Behavior

  • No AVX warning (should use arm64 Bun)
  • Reasonable number of node processes that clean up on exit

Screenshots

[Activity Monitor showing 40+ node processes at ~70MB each]

Workaround

pkill -9 node to kill zombie processes, but they respawn on next launch

<img width="1148" height="1047" alt="Image" src="https://github.com/user-attachments/assets/5f2fef66-a983-4980-b00d-b62670de903e" />

<img width="1086" height="710" alt="Image" src="https://github.com/user-attachments/assets/476ec83f-8789-4e61-947e-4cb8646c784a" />

What Should Happen?

CC shouldn't crash and cause oom to my computer, tried to reinstall and still doesn't workmacOS on M4 Max
Claude Code v2.0.76
Node processes spawning on startup (40+ at ~70MB each = 3GB+ leak)
The AVX/Bun warning (suggests wrong architecture build)

Error Messages/Logs

Steps to Reproduce

macOS on M4 Max
Claude Code v2.0.76
Node processes spawning on startup (40+ at ~70MB each = 3GB+ leak)
The AVX/Bun warning (suggests wrong architecture build)

Claude Model

Opus

Is this a regression?

Yes, this worked in a previous version

Last Working Version

_No response_

Claude Code Version

2.0.76

Platform

Anthropic API

Operating System

macOS

Terminal/Shell

Cursor

Additional Information

_No response_

View original on GitHub ↗

This issue has 7 comments on GitHub. Read the full discussion on GitHub ↗