ClawBox Pro

Professional AI Workstation

ClawBox Pro doubles the memory and storage for professionals who need to run larger models, handle concurrent requests, and deploy production AI workloads. The go-to choice for teams, studios, and power users.

Hardware Specifications

• AI-Optimized SoC with enhanced NPU + GPU
• 32GB Unified Memory — run 7B-30B models, or multiple 7B models simultaneously
• 512GB NVMe Storage — store 15-20 models locally
• 2.5GbE Ethernet + Wi-Fi 6E
• 25-45W Power — near-silent operation
• Rack-mountable with included brackets

Software Included

• Everything in Starter, plus:
• Advanced AI router with priority queues
• Multi-user access control (RBAC)
• Prometheus metrics exporter
• Syslog integration for enterprise monitoring
• API rate limiting and usage dashboards

Performance

• Llama 3.1 8B: ~38 tok/s (single) / ~25 tok/s (4 concurrent)
• Llama 3.1 70B (Q4): ~6.8 tok/s generation
• Multi-model: Run 3 different 7B models simultaneously

Perfect For

Professional developers, small-to-medium teams, startups, and any organization needing reliable, always-on local AI. Ideal for production APIs, team-shared AI assistants, and compliance-sensitive deployments.

Open Source: 100% Apache 2.0 licensed. View source on GitHub.
Warranty: 2-year hardware warranty. 30-day returns for unopened units.
Shipping: Free US shipping. Ships within 5-7 business days.