GPUs are open to everyone!
As of right now, here are the GPUs you can use and their cost:
| Model name | Fly name | VRAM | Cost per hour |
|---|---|---|---|
| Ampere A100 80 GB | a100-80gb |
80 GB | $3.50 |
| Ampere A100 40 GB | a100-40gb |
40 GB | $2.50 |
| Lovelace L40S | l40s |
48 GB | $2.50 |
If you need some ideas to get started with GPUs, see here:
- Deploy Your Own (Not) MidJourney Bot On Fly GPUs
- Scaling Large Language Models to zero with Ollama
- Transcribing on Fly GPU Machines
Have fun!