L1 AI Pod
The standard for on-prem inference. A self-contained, acoustically damped cluster designed for office environments.
Hardware
Interactive Component Explorer
Explore each component of the L1 AI Pod in 3D. Hover to highlight, click for detailed specifications.
4× NVIDIA H100 PCIe
gpu
320GB HBM3 Total | 15,832 TFLOPS FP16
2× AMD EPYC 7713
cpu
128 Cores | 256 Threads | 512MB L3
512GB DDR4 ECC
memory
8× 64GB DIMMs | 3200MHz | Registered
8TB NVMe SSD
storage
7,000 MB/s Read | PCIe 4.0 x4
Dual 100GbE
network
Mellanox ConnectX-6 | RDMA Support
Power Delivery System
power
208V/30A | 2× 2000W Redundant PSU
Features
Why Choose L1 Pod
Self-contained cluster designed for office environments
Acoustically damped for minimal noise disruption
Pre-configured software stack ready for deployment
Full electrical and HVAC integration included
Ongoing hardware support and maintenance
Air-gapped security capabilities
Sub-millisecond local inference latency
Scalable to multiple units
Ready to Deploy Your L1 Pod?
Schedule your 60-minute feasibility survey to see if your office is ready for on-prem AI infrastructure.