Lambda Blade

Enterprise

Drastically improve TensorFlow and model training performance.

Featuring up to 8x 16GB GPUs, Lamba Blade is the most powerful workstation we sell. It's built for enterprise scale.

This is the enterprise solution needed for parallelization and cutting-edge deep learning performance. Experience how profitable of an investment deep learning hardware can be for your team.

$16,180
* Does not include shipping
Ships in 7-10 business days
Get help buying:

Specs

GPU
4x 1080
Ti

GeForce GTX 1080 Ti, GeForce GTX Titan Xp, or Tesla P100. Pascal architecture

Processor
2x Intel Xeon

2x Intel Xeon® Processor E5-2650, 12 core, 2.20 GHz

Storage
1TB
SSD

Configuration options available

RAM
128GB
ECC

Configurable up to 512GB.

Our GPUs are
3X
faster than previous-generation graphic cards
Total performance up to
2X
faster than Lambda QUAD
Equipped with
32X
more RAM than base model

Lambda Blade

Lambda Customers

Relative Performance

BLADE is our enterprise rackmount solution. Provide your team with the fastest GPU deep learning machine on the market. The BLADE can be configured with either Pascal 1080Ti or Pascal P100 GPUs and can deliver training performance that is 400x faster than a CPU server. Accelerate your datacenter with BLADE GPU compute.

It takes a massive amount of computing performance to train the sophisticated deep neural networks that power new applications in the sciences and financial markets. Training can take days to weeks on even the fastest supercomputers. With GPU acceleration, neural net training is 10-20 times faster than with CPUs. That means training is reduced from days or even weeks to just hours. This lets researchers and data scientists build larger, more sophisticated neural nets.

Frequently Asked Questions

How much RAM do I need? 64GB or 128GB?

In general, we recommend that you have at least as much system RAM as you do GPU RAM. Thus, with a 4 GPU setup we recommend 64GB and with an 8 GPU setup we recommend 128GB.

What about the Titan Xp — isn't that better?

No, the 1080Ti is only 8% slower but is half the price. On a price / performance basis the 1080Ti is the best GPU for Deep Learning.

We pre-configure every workstation

Our high performance workstations are built specifically for Deep Learning. They're pre-configured with the libraries you need. And they're backed by professional support.

Deep learning hardware built by researchers

We build deep learning machines for researchers and corporations.

Our salespeople are engineers. When you chat with us, there's no sales talk. We're a small team of experts who take pride in our machines.

Call us. Tell us what you're working on, and we'll talk you through your options.

Ask about our support: If a hardware component fails, we understand your machine must be fixed immediately. That’s why we offer up to three years of on-site support. If something goes wrong, we'll have an expert fix it.

Call us:
1 (650) 479-5530
or, Configure yours