Node 1 — OpenClaw + Ollama MLX
The Intelligence Hub. Orchestration + LLM Inference with Apple Silicon optimization.
Hardware
- Machine: Mac Studio M4 Max
- RAM: 36GB Unified Memory
- Storage: Internal SSD
Software Stack
OpenClaw
AI orchestration and automation layer. Coordinates agent workflows, manages memory, and handles complex multi-step tasks.
Ollama MLX
Local LLM inference engine with Apple Silicon Metal GPU acceleration. Runs MLX models with native performance.
Models
Local Models (MLX)
- Qwen 2.5 — Multilingual instruction following
Cloud Models
- MiniMax — Cloud inference for complex reasoning tasks
Capabilities
- Local LLM inference with Metal GPU acceleration
- Cloud inference integration (MiniMax)
- Agentic workflows with OpenClaw orchestration
- Multi-step task automation
- Memory and context management
- Integration with Node 2 for distributed workloads