You can spin up up to 64+ Multi-Node H100 GPUs on Prime Intellect On-demand.

​
Step-by-Step Guide

1

Go to the Megacluster tab inside the platform

2

Choose your preferred configration of 16-64+ H100 GPUs and click 'Deploy Cluster'

3

Wait for the cluster to be deployed. You'll receive an email once it's up.

4

You'll receive one public IP for every node (8xH100) you deployed.

You can use these public IPs to SSH into your nodes and start running your multi-node use case.

To see how to run Megatron-Deepspeed, Huggingface Accelerate, Torch FSDP, and other multi-node use cases, refer to our Tutorials:

Run Torch FSDP

Run Megatron-Deepspeed

Run Huggingface Accelerate

Run LLaMA 405B in BF16

You can spin up up to 64+ Multi-Node H100 GPUs on Prime Intellect On-demand.

​
Step-by-Step Guide

1

Go to the Megacluster tab inside the platform

2

Choose your preferred configration of 16-64+ H100 GPUs and click 'Deploy Cluster'

3

Wait for the cluster to be deployed. You'll receive an email once it's up.

4

You'll receive one public IP for every node (8xH100) you deployed.

You can use these public IPs to SSH into your nodes and start running your multi-node use case.

To see how to run Megatron-Deepspeed, Huggingface Accelerate, Torch FSDP, and other multi-node use cases, refer to our Tutorials:

Run Torch FSDP

Run Megatron-Deepspeed

Run Huggingface Accelerate

Run LLaMA 405B in BF16