Vast.ai Python SDK & CLI
The official Vast.ai Python package — provides both the CLI and SDK for managing Vast.ai GPU cloud resources, plus a serverless client for endpoint inference.
Install
Linux, macOS, or WSL — install the CLI with no Python required:
curl -fsSL https://vast.ai/install.sh | bash
This installs vastai into an isolated managed runtime under
~/.local/share/vastai, decoupled from any system or project Python. Update
anytime with vastai update, or pin/roll back with vastai update --version X.
Windows, or using the Python SDK — install from PyPI instead:
pip install vastai
Note:
pip install vastai-sdkalso works and installs the same package. Both package names are supported for backward compatibility.
Quickstart
-
Get your API key from https://cloud.vast.ai/manage-keys/
-
Set your API key:
vastai set api-key YOUR_API_KEY
- Test a search:
vastai search offers --limit 3
You should see a short list of available GPU offers.
CLI Usage
The vastai command provides full access to the Vast.ai platform from your terminal:
vastai search offers 'gpu_name=RTX_4090 num_gpus>=4'
vastai create instance 12345 --image pytorch/pytorch --disk 32 --ssh --direct
vastai show instances
vastai stop instance 12345
vastai destroy instance 12345
Run vastai --help for a full list of commands. You can also use --help on any subcommand:
vastai search offers --help
vastai create instance --help
SDK Usage
from vastai import VastAI
vast = VastAI() # uses VAST_API_KEY env var, or pass api_key="..."
vast.search_offers(query='gpu_name=RTX_4090 num_gpus>=4')
vast.show_instances()
vast.start_instance(id=12345)
vast.stop_instance(id=12345)
Use help(vast.search_offers) to view documentation for any method.
Migrating from
vastai-sdk? The old import still works:from vastai_sdk import VastAI
Using the Serverless Client
- Create the client
from vastai import Serverless
serverless = Serverless() # or, Serverless("YOUR_API_KEY")
- Get an endpoint
endpoint = await serverless.get_endpoint("my-endpoint")
- Make a request
request_body = {
"model": "Qwen/Qwen3-8B",
"prompt" : "Who are you?",
"max_tokens" : 100,
"temperature" : 0.7
}
response = await serverless.request("/v1/completions", request_body)
- Read the response
text = response["response"]["choices"][0]["text"]
print(text)
Find more examples in the examples/ directory.
Tab Completion
Tab completion is supported in Bash and Zsh via argcomplete (installed automatically). To enable it:
activate-global-python-argcomplete
Or for a single session:
eval "$(register-python-argcomplete vastai)"
AI Agents
Vast.ai has a skill for AI coding agents (Claude Code, Cursor, Windsurf, Codex, etc.):
npx skills add vast-ai/vast-cli
This installs the Vast.ai skill so your agent can search offers, create instances, and manage GPU workflows directly. See CLI SKILL.md or SDK SKILL.md for the full reference.
Contributing
This repository is open source. If you find a bug, please open an issue. PRs are welcome.
3 maintainers
Release files for vastai 1.5.5
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| vastai-1.5.5.tar.gz | 264.8 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| vastai-1.5.5-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 566.9 kB
Release files / vastai-1.5.5.tar.gz
| Download URL | vastai-1.5.5.tar.gz |
|---|---|
| Size | 264.8 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
cd1f7b50dff7277a2119b2240a752a2ab785d818ae33815771c2a210dff78d4e
|
|
BLAKE2b-256 checksum How to use checksums |
4c215e1e50ed67c5de6b0e44ea68f83922c49751d4fb1f18c4945a01ee56abb6
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
poetry/2.4.1 CPython/3.12.13 Linux/6.17.0-1022-azure
|
Release files / vastai-1.5.5-py3-none-any.whl
| Download URL | vastai-1.5.5-py3-none-any.whl |
|---|---|
| Size | 302.1 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
cbc80ead74256f03f09dd575ee0d16d95d21326a9c70feba186a99862ad2ccf4
|
|
BLAKE2b-256 checksum How to use checksums |
d861964e3d610a07567bc1f70408501936fd6ad02c15735a53565faf57c27dc3
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
poetry/2.4.1 CPython/3.12.13 Linux/6.17.0-1022-azure
|
