v1.0.3

v1.0.3 - Enhanced Reliability & User Experience

🎯 Major Improvements

Auto-Launch Claude Code

• claude-* aliases now automatically launch Claude Code after starting the backend, matching the codex-* workflow
• No more manual claudel command needed - just run claude-local-glm-9b and you're ready to code!

Robust Model Download Handling

• Added comprehensive model validation for sharded models (checks all parts are present)
• Detects incomplete downloads and automatically cleans up corrupted model caches
• Waits for models to be fully loaded into memory before proceeding
• Prevents launching CLI tools with unready backends

Better Process Management

• Ctrl-C now properly cleans up background processes (no more orphaned downloads)
• PID files saved immediately to enable proper cleanup on interruption
• Improved child process termination (kills download subprocesses)

📚 Documentation

• Added ASCII architecture diagram to README for quick understanding
• Created comprehensive CLAUDE.md with setup instructions, architecture overview, and troubleshooting
• Improved Python version compatibility checks in setup

🐛 Bug Fixes

• Fixed default Codex command to avoid unintended profile overrides
• Reduced verbose output in status.sh for cleaner display
• Better alignment between codex-models and claude-models output

🔧 Technical Details

Validation Improvements:

• Checks model.safetensors.index.json for sharded models
• Verifies all weight files are present and non-empty
• Validates tokenizer files exist
• Tests actual model functionality before proceeding

Signal Handling:

• Trap SIGINT/SIGTERM for graceful cleanup
• Kills both parent and child processes
• Removes PID files on exit

Full Changelog: 1.0.2...1.0.3

v1.0.2

v1.0.2 - Codex CLI Integration

✨ New Features

• Codex CLI support - Use any model with Codex CLI via new codex-* aliases
• Auto-generated profiles - ./scripts/setup-codex.sh creates Codex profiles automatically
• SEED-OSS 36B model - New LM Studio configuration (36B params, 512K context)

🔧 Improvements

• Better error handling - LM Studio startup now tests actual model functionality
• Cleaner documentation - Simplified model examples, updated project structure
• Repository rename - Now called "ai-cli-switchboard" to reflect broader scope

Works with both Claude Code and Codex CLI using the same backend configurations.

v1.0.1

Updated README

v1.0.0

v1.0.0 - Initial Release

Overview

A framework for using Claude Code with any local or remote LLM on Apple Silicon.

Features

• Support for 20+ models across 4 runner types (MLX, LM Studio, LiteLLM, Z.AI)
• Dynamic alias generation from config files
• Unified start/stop/status management
• One-script setup with ./setup.sh

Supported Models (* recommended)

• Remote: *DeepSeek-R1, *Gemini 2.5, *GLM-4.5, OpenRouter models
• Local MLX: GLM-4-9B/32B, DeepSeek-V2.5, FuseO1, Gemma 2B
• LM Studio: Llama 3 Groq, Qwen3 variants, Phi-4, DeepSeek R1

Requirements

• macOS with Apple Silicon
• Python 3.10+
• Claude Code CLI

Getting Started

1. Run ./setup.sh
2. Add API keys to .env
3. Set up aliases: ./scripts/setup-aliases.sh && source ai-aliases.sh
4. List available models: claude-models
5. Start a model: e.g. claude-remote-deepseek
6. Use Claude Code: claudel
7. Add your own configs and experiement

Full Changelog: https://github.com/PatchedReality/claude-multi-model/commits/1.0