What you can do with Jarvislabs.ai?
- Launch an GPU/CPU powered instance.
- Pause the instance.
- Resume the instance.
- Resume it by changing the GPU type, scaling GPUs or storage.
- SSH to an instance.
- Run from your own docker image.
- Deploy your model using Flask/FastAPI
Pick a modern Nvidia GPU to power your Deep Learning
We support the latest GPU based on Nvidia Ampere architecture (2022), which are usually 1.3X - 3X times faster than GPUs based on Turing Architecture. A100 is the fastest GPU, followed by A6000 and A5000.
| GPU | vCPUS | Ram | Reserved | Spot | Weekly | Monthly |
|---|---|---|---|---|---|---|
| A100 - 40GB | 7 | 32GB | $2.39/hr | $0.99/hr | $360/week | $1200/month |
| A6000 - 48GB | 7 | 32GB | $1.79/hr | $0.79/hr | $270/week | $900/month |
| A5000 - 24GB | 7 | 32GB | $1.29/hr | $0.59/hr | $195/week | $650/month |
You can also choose GPU based on Nvidia Turing Architecture (2019), which offers similar performance like Nvidia V100 GPU.
| GPU | vCPUS | Ram | Reserved | Spot | Weekly | Monthly |
|---|---|---|---|---|---|---|
| RTX 6000 - 24GB | 7 | 32GB | $0.99/hr | $0.39/hr | $149/week | $500/month |
| RTX 5000 - 16GB | 7 | 32GB | $0.49/hr | $0.19/hr | $75/week | $245/month |
Need any assistance, say 👋 Hi to the team using the chat option available on the website or drop us an email to hello@jarvislabs.ai