https://github.com/dstackai/gpuhunt
GPU prices aggregator for cloud providers
https://github.com/dstackai/gpuhunt
Last synced: 9 months ago
JSON representation
GPU prices aggregator for cloud providers
- Host: GitHub
- URL: https://github.com/dstackai/gpuhunt
- Owner: dstackai
- License: mpl-2.0
- Created: 2023-08-23T06:53:10.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2025-04-30T11:41:36.000Z (9 months ago)
- Last Synced: 2025-05-11T17:58:26.074Z (9 months ago)
- Language: Python
- Size: 320 KB
- Stars: 37
- Watchers: 6
- Forks: 14
- Open Issues: 10
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
[](https://pypi.org/project/gpuhunt/)
Easy access to GPU pricing data for major cloud providers: AWS, Azure, GCP, etc.
The catalog includes details about prices, locations, CPUs, RAM, GPUs, and spots (interruptible instances).
## Usage
```python
import gpuhunt
items = gpuhunt.query(
min_memory=16,
min_cpu=8,
min_gpu_count=1,
max_price=1.0,
)
print(*items, sep="\n")
```
List of all available filters:
* `provider`: name of the provider to filter by. If not specified, all providers will be used. One or many
* `cpu_arch`: CPU architecture, one of: `x86`, `arm`
* `min_cpu`: minimum number of CPUs
* `max_cpu`: maximum number of CPUs
* `min_memory`: minimum amount of RAM in GB
* `max_memory`: maximum amount of RAM in GB
* `min_gpu_count`: minimum number of GPUs
* `max_gpu_count`: maximum number of GPUs
* `gpu_vendor`: GPU/accelerator vendor, one of: `nvidia`, `amd`, `google`, `intel`
* `gpu_name`: name of the GPU to filter by. If not specified, all GPUs will be used. One or many
* `min_gpu_memory`: minimum amount of GPU VRAM in GB for each GPU
* `max_gpu_memory`: maximum amount of GPU VRAM in GB for each GPU
* `min_total_gpu_memory`: minimum amount of GPU VRAM in GB for all GPUs combined
* `max_total_gpu_memory`: maximum amount of GPU VRAM in GB for all GPUs combined
* `min_disk_size`: minimum disk size in GB (not fully supported)
* `max_disk_size`: maximum disk size in GB (not fully supported)
* `min_price`: minimum price per hour in USD
* `max_price`: maximum price per hour in USD
* `min_compute_capability`: minimum compute capability of the GPU
* `max_compute_capability`: maximum compute capability of the GPU
* `spot`: if `False`, only ondemand offers will be returned. If `True`, only spot offers will be returned
## Advanced usage
```python
from gpuhunt import Catalog
catalog = Catalog()
catalog.load(version="20240508")
items = catalog.query()
print(*items, sep="\n")
```
## Supported providers
* AWS
* Azure
* Cudo Compute
* DataCrunch
* GCP
* LambdaLabs
* Nebius
* OCI
* RunPod
* TensorDock
* Vast AI
* Vultr
## See also
* [dstack](https://github.com/dstackai/dstack)