Local AI workstation for private inference with up to 192 GB of unified memory and 32 GB Radeon GPU
Zero-retention inference for frontier models
Stage
Not given
Startup
Founded
Not given
Not given
Based in
Not given
MX
Pricing model
Paid
Freemium
Pricing
Developer: starting at $5,999 USD (128 GB memory, 2 TB storage). Memory upgrade to 192 GB adds $2,200. Storage upgrade to 4 TB adds $250. Cluster kit $799 per machine. Business pricing available by quote. Worldwide shipping, import duties and taxes outside US included. No sales tax in US.
Pay as you go from $10. Monthly subscriptions: Run $19/month, Run Plus $99/month (5× allowance), Run Max $199/month (10× allowance). Model pricing per million tokens varies: qwen3.8-max $2.00 input/$6.00 output, claude-haiku-5-5 $0.10 input/$0.50 output, claude-sonnet-5 $2.00 input/$10.00 output, cl
Key features
Run DeepSeek V4 Flash and Qwen 27B models
Browser-based setup over encrypted Bluetooth
Private OpenAI-compatible API
Lucebox Mix: run two models simultaneously
Headless Ubuntu Server with pre-installed inference engine
Optional Tailscale integration for remote access
Optional cluster kit with 100 GbE networking
Zero-retention guarantees with signed cryptographic receipts for every API call
OpenAI-compatible API endpoint
Web-based chat interface with encrypted conversations
Access to 15+ frontier models including Claude, Gemini, Qwen, DeepSeek
Python SDK and R package support
Receipts show token usage, cost, model, provider and retention basis
No data storage on offrec infrastructure
Models run on managed third-party providers
What makes it different
No per-token billing or cloud dependencies
1.5x faster decode speed than DGX Spark
Hardware and software co-designed for optimal performance
Completely private: runs offline with no cloud proxy
Open source inference engine with tuned models
Every call returns a cryptographically signed receipt proving zero retention
No need to negotiate zero-retention terms with each provider individually
Unified access to frontier models from multiple labs via single API key
Offrec owns no infrastructure; models run on managed providers
Verifiable accountability through transparent receipt system
Lucebox
offrec