{"categories":[{"label":"Distributed Computing","url":"https://skillfed.io/packages/category/system-distributed-computing"}],"enrichment":{"capability":"Provides Python bindings to the Vast.ai API for searching GPU compute offers, managing instances, and accessing serverless inference endpoints.","skillfed_tags":["gpu-compute","cloud-api"],"use_cases":["Search and filter GPU compute offers by hardware specs, price, or availability.","Start, stop, and monitor GPU instances programmatically for batch training or inference workloads.","Call serverless inference endpoints for LLM completions or other model serving."],"what_it_does":"vastai-sdk is a Python client library for the Vast.ai GPU compute marketplace. It wraps the Vast.ai REST API to let you search available GPU instances, manage your rented compute, and interact with serverless inference endpoints. The package supports both synchronous instance management (searching, starting, stopping instances) and asynchronous serverless client operations for making inference requests.\n\nThe package is officially deprecated and has been merged into a unified vastai package; installing vastai-sdk now installs the same underlying package. It remains available as a compatibility wrapper for existing code. The library requires Python 3.9 or later and depends only on the vastai runtime package.","worth_installing":"Yes, but with a caveat: install this only if you have existing code that imports from vastai-sdk. For new projects, use the vastai package instead. The deprecation signals this wrapper exists for backward compatibility only, though the underlying vastai package is actively maintained."},"id":"vastai-sdk","links":{"html":"https://skillfed.io/packages/vastai-sdk","md":"https://skillfed.io/packages/vastai-sdk.md","pypi":"https://pypi.org/project/vastai-sdk/"},"maintenance":{"status":"active"},"meta":{"latest_release":"2026-08-12","license_spdx":null,"license_treatment":"unclear","name":"vastai-sdk","python_support":"supports_current","summary":"DEPRECATED \u2014 use 'pip install vastai' instead. This package is a compatibility wrapper that installs vastai."},"popularity":{"monthly_downloads":3218711,"position":2696,"tier":"top_5000"},"security":{"n_vulnerabilities":0},"version":"1.5.4"}
