The world's fastest server for AI research

5000+ research groups trust Lambda






Easy system administration
Our GPU servers comes with Lambda Stack, which includes frameworks like TensorFlow and PyTorch. Lambda Stack makes upgrading these frameworks easy.




Technical specifications

GPUs
4 NVIDIA® A100 SXM4 GPUs (40 GB or 80 GB) · NVLink and PCIe 4.0 GPU-to-GPU...
GPUs
- Up to 4 NVIDIA® A100 SXM4 GPUs (40 GB or 80 GB)
- NVLink and PCIe 4.0 GPU-to-GPU interconnect
Processors
2 AMD EPYC™ 7003 series CPUs · Up to 64 cores, 256 threads, 256 MB cache per processer...
Processors
2 AMD EPYC™ 7003 series CPUs. Options include:
- AMD EPYC 7313: 16 cores, 32 threads, 3.0 GHz base, 3.70 GHz boost, 128 MB cache
- AMD EPYC 7413: 24 cores, 48 threads, 2.65 GHz base, 3.60 GHz boost, 128 MB cache
- AMD EPYC 7513: 32 cores, 64 threads, 2.60 GHz base, 3.65 GHz boost, 128 MB cache
- AMD EPYC 7643: 48 cores, 96 threads, 2.30 GHz base, 3.60 GHz boost, 256 MB cache
- AMD EPYC 7663: 56 cores, 112 threads, 2.0 GHz base, 3.50 GHz boost, 256 MB cache
- AMD EPYC 7713: 64 cores, 128 threads, 2.0 GHz base, 3.675 GHz boost, 256 MB cache
- AMD EPYC 7763: 64 cores, 128 threads, 2.45 GHz base, 3.50 GHz boost, 256 MB cache
System memory
Up to 4096 GB of 3200 MHz DDR4 ECC RAM in 32 DIMM slots · 8 memory channels per...
System memory
- Up to 4096 GB of 3200 MHz DDR4 ECC RAM in 32 DIMM slots
- 8 channels per processor
Operating system drive
Up to 15.36 TB of storage via one 2.5" NVME SSD (PCIe 3.x or 4.0)
Operating system drive
Up to 15.36 TB of storage via one 2.5" NVME SSD (PCIe 3.x or 4.0)
Additional storage drives
Up to 46.08 TB of additional storage via 3 hot-swappable 2.5" NVMe SSDs (PCIe 3.x or 4.0)...
Additional storage drives
- Up to 46.08 TB of additional storage via 3 hot-swappable 2.5" NVMe SSDs (PCIe 3.x or 4.0)
- Designed to store large data sets and model checkpoints
Networking
2 RJ45 10 Gbps (aggregate) BASE-T LAN ports · 1 RJ45 1 Gbps BASE-T LAN out-of-band...
Networking
Built-in networking:
- 2 RJ45 10 Gbps (aggregate) BASE-T LAN ports
- 1 RJ45 1 Gbps BASE-T LAN out-of-band management (IPMI) port
Up to 4 high-speed NICs for GPUDirect RDMA. Options include:
- 200 Gbps InfiniBand NVIDIA HDR ConnectX-6 VPI Adapter Card, 2x QSFP56, PCIe 4.0 x16
- 200 Gbps InfiniBand NVIDIA HDR ConnectX-6 VPI Adapter Card, 1x QSFP56, PCIe 4.0 x16
- 100 Gbps InfiniBand NVIDIA HDR100 ConnectX-6 VPI Adapter Card, 1x QSFP56, PCIe 4.0 x16
- 100 Gbps InfiniBand NVIDIA HDR100 ConnectX-6 VPI Adapter Card, 2x QSFP56, PCIe 4.0 x16
- 200 Gbps Ethernet NVIDIA ConnectX-6 EN Adapter Card, 2x QSFP56, PCIe 4.0 x16
- 200 Gbps Ethernet NVIDIA ConnectX-6 Dx EN Adapter Card, 1x QSFP56, PCIe 4.0 x16
- 100 Gbps Ethernet NVIDIA ConnectX-6 EN Adapter Card, 2x QSFP56, PCIe 4.0 x16
- 100 Gbps Ethernet NVIDIA ConnectX-6 Dx EN Adapter Card, 1x QSFP56, PCIe 4.0 x16
Power
Two 2200 watt 80 PLUS™ Platinum PSUs · 2 + 0 redundancy...
Power
- Two 2200 watt 80 PLUS™ Platinum PSUs
- 2 + 0 redundancy
AC input:
- 220-240Vac / 11-10 amperes / 50-60 Hz
Max output per PSU:
- 220-240 Vac: 2200 watts
Front I/O
Power button · Reset button · Power LED · Information LED · Power supply failure LED · 2 LAN...
Front I/O
- Power button
- Reset button
- Power LED
- Information LED
- Power supply failure LED
- 2 LAN activity LEDs, one for each RJ45 10 Gbps (aggregate) BASE-T LAN port
- Storage drive activity LED
Rear I/O
UID button with LED · VGA port · 2 USB 3.0 ports · 2 RJ45 10 Gbps (aggregate) BASE-T LAN...
Rear I/O
- UID button with LED
- VGA port
- 2 USB 3.0 ports
- 2 RJ45 10 Gbps (aggregate) BASE-T LAN ports
- 1 RJ45 1 Gbps BASE-T LAN out-of-band management (IPMI) port
Accessories
Rackmounting kit · 4 configurable power cables
Accessories
- Rackmounting kit
- 4 configurable power cables
Dimensions & weight
2 rack units (2U) · 17.2 in (width) x 3.5 in (height) x 32.7 in (depth) · 88.5 lbs with packaging...
Dimensions & weight
- Form factor: 2U rackmount
- Width: 17.2 inches (437 mm)
- Height: 3.5 inches (89 mm)
- Depth: 32.7 inches (830.3 mm)
- Server weight: 78.5 lbs (35.6 kg)
- Rackmounting kit weight: 5 lbs (2.3 kg)
- Total weight with packaging: 88.5 lbs (40.1 kg)

GPUs
8 NVIDIA® A100 SXM4 GPUs (40 GB or 80 GB) · NVLINK™ and NVIDIA® NVSwitch™...
GPUs
- 8 NVIDIA® A100 SXM4 GPUs (40 GB or 80 GB)
- NVLINK™ and NVIDIA® NVSwitch™ fabric
Processors
2 AMD EPYC™ 7003 series CPUs · Up to 64 cores, 256 threads, 256 MB cache per processer...
Processors
2 AMD EPYC™ 7003 series CPUs. Options include:
- AMD EPYC 7313: 16 cores, 32 threads, 3.0 GHz base, 3.70 GHz boost, 128 MB cache, PCIe 4.0
- AMD EPYC 7413: 24 cores, 48 threads, 2.65 GHz base, 3.60 GHz boost, 128 MB cache, PCIe 4.0
- AMD EPYC 7513: 32 cores, 64 threads, 2.60 GHz base, 3.65 GHz boost, 128 MB cache, PCIe 4.0
- AMD EPYC 7643: 48 cores, 96 threads, 2.30 GHz base, 3.60 GHz boost, 256 MB cache, PCIe 4.0
- AMD EPYC 7663: 56 cores, 112 threads, 2.0 GHz base, 3.50 GHz boost, 256 MB cache, PCIe 4.0
- AMD EPYC 7713: 64 cores, 128 threads, 2.0 GHz base, 3.675 GHz boost, 256 MB cache, PCIe 4.0
- AMD EPYC 7763: 64 cores, 128 threads, 2.45 GHz base, 3.50 GHz boost, 256 MB cache, PCIe 4.0
System memory
Up to 4096 GB of 3200 MHz DDR4 ECC RAM in 32 DIMM slots · 8 memory channels per processor...
System memory
- Up to 4096 GB of 3200 MHz DDR4 ECC RAM in 32 DIMM slots
- 8 channels per processor
Operating system drive
Up to 15.36 TB of storage via one 2.5" NVME SSD (PCIe 3.x or PCIe 4.0)
Operating system drive
Up to 15.36 TB of storage via one 2.5" NVME SSD (PCIe 3.x or PCIe 4.0)
Additional storage drives
Up to 138.24 TB of additional storage via 9 NVMe SSDs (PCIe 3.x or 4.0) · Designed to store...
Additional storage drives
- Up to 138.24 TB of additional storage via 9 NVMe SSDs (PCIe 3.x or 4.0)
- Designed to store large data sets and model checkpoints
Networking
2 RJ45 10 Gbps BASE-T LAN ports · 1 RJ45 1 Gbps BASE-T LAN out-of-band management (IPMI)...
Networking
Built-in networking:
- 2 RJ45 10 Gbps BASE-T LAN ports
- 1 RJ45 1 Gbps BASE-T LAN out-of-band management (IPMI) port
Up to 8 high-speed NICs for GPUDirect RDMA. Options include:
- 200 Gbps InfiniBand NVIDIA HDR ConnectX-6 VPI Adapter Card, 2x QSFP56, PCIe 4.0 x16
- 200 Gbps InfiniBand NVIDIA HDR ConnectX-6 VPI Adapter Card, 1x QSFP56, PCIe 4.0 x16
- 100 Gbps InfiniBand NVIDIA HDR100 ConnectX-6 VPI Adapter Card, 1x QSFP56, PCIe 4.0 x16
- 100 Gbps InfiniBand NVIDIA HDR100 ConnectX-6 VPI Adapter Card, 2x QSFP56, PCIe 4.0 x16
- 200 Gbps Ethernet NVIDIA ConnectX-6 EN Adapter Card, 2x QSFP56, PCIe 4.0 x16
- 200 Gbps Ethernet NVIDIA ConnectX-6 Dx EN Adapter Card, 1x QSFP56, PCIe 4.0 x16
- 100 Gbps Ethernet NVIDIA ConnectX-6 EN Adapter Card, 2x QSFP56, PCIe 4.0 x16
- 100 Gbps Ethernet NVIDIA ConnectX-6 Dx EN Adapter Card, 1x QSFP56, PCIe 4.0 x16
Power
4 hot-swappable 3000 watt 80 PLUS™ Titanium power supplies · 2 + 2 redundancy...
Power
- 4 hot-swappable 3000 watt 80 PLUS™ Titanium PSUs
- 2 + 2 redundancy
AC input:
- 200-207 Vac / 16-15.7 amperes / 50-60 Hz
- 207.1-240 Vac / 16-14.5 amperes / 50-60 Hz
Max output per PSU:
- 200-207 Vac: 2880 watts
- 207.1–240 Vac: 3000 watts
Front I/O
Power button · Reset button · Power LED · UID button with LED · Information LED · 1 RJ45...
Front I/O
- Power button
- Reset button
- Power LED
- UID button with LED
- Information LED
- 1 RJ45 1 Gbps BASE-T LAN out-of-band management (IPMI) port
- LAN activity LED for out-of-band management (IPMI) port
Rear I/O
2 RJ45 10 Gbps BASE-T LAN ports
Rear I/O
2 RJ45 10 Gbps BASE-T LAN ports
Accessories
Rackmounting kit · 4 C19 TO C20 power cables (4 foot)
Accessories
- Rackmounting kit
- 4 C19 TO C20 power cables (4 foot)
Dimensions & weight
4 rack units (2U) · 17.6 in (width) x 6.9 in (height) x 35.4. in (depth) · 225 lbs with packaging...
Dimensions & weight
- Form factor: 4U rackmount
- Width: 17.6 inches (446 mm)
- Height: 6.9 inches (174 mm)
- Depth: 35.4 inches (900 mm)
- Server weight: 166 lbs (75.3 kg)
- Rackmounting kit weight: 5 lbs (2.3 kg)
- Total weight with packaging: 225 lbs (102.1 kg)
Our published research
View all researchMultiple Pairwise Ranking Networks for Personalized Video Summarization
We propose a model for personalized video summaries by conditioning the summarization process with predefined categorical user labels.
HoloGAN: Unsupervised Learning of 3D Representations from Natural Images
We propose a novel generative adversarial network (GAN) for the task of unsupervised learning of 3D representations from natural images.
RenderNet: A Deep ConvNet for Differentiable Rendering from 3D Shapes
We present a differentiable rendering convolutional network with a novel projection unit that can render 2D images from 3D shapes.
Adversarial Monte Carlo Denoising with Conditioned Aux. Feature Modulation
We demonstrate that GANs can help denoiser networks produce more realistic high-frequency details and global illumination.