1. Home
  2. Companies
  3. Vast.ai
Vast.ai logoVA

Vast.ai

About

Vast.ai operates a cloud GPU marketplace founded in 2018, connecting data centers and professionals running its hosting software with users who need GPU compute resources. The platform is designed for workloads in AI, machine learning, deep learning, rendering, and data processing. It functions as a peer-to-peer marketplace, offering access to over 10,000 GPUs across a network of 40 secure data centers worldwide.

The company's marketplace model positions it as a cost-effective alternative to traditional cloud providers, with stated savings of up to 80% on GPU rentals. Vast.ai's stated mission is to organize, optimize, and orient the world's computation, with a broader vision related to making life substrate-independent through artificial intelligence.

Similar companies

NVIDIA Corporation logoNC

NVIDIA Corporation

NVIDIA designs and manufactures GPUs and accelerated computing platforms for applications in AI, high-performance computing, gaming, autonomous vehicles, and scientific research.

4 jobs
Hyperbolic Labs logoHL

Hyperbolic Labs

Hyperbolic Labs operates an Open-Access AI Cloud providing a GPU marketplace and serverless AI inference service for developers.

1 job
Runpod logoRU

Runpod

Runpod builds an AI infrastructure platform providing GPU compute, model training, inference, and agent orchestration for over 500,000 developers worldwide.

1 job
Prime Intellect logoPI

Prime Intellect

Prime Intellect builds decentralized AI infrastructure, operating a compute marketplace for up to 5,000 GPUs and providing an open stack for collaborative model training and reinforcement learning.

1 job
CoreWeave logoCO

CoreWeave

CoreWeave is a specialized cloud platform providing GPU infrastructure for training and deploying large AI models, with data centers in the US and Europe.

Hammerspace logoHA

Hammerspace

Hammerspace provides a Global Data Environment platform that unifies unstructured data across edge, core, and multi-cloud to eliminate silos and accelerate AI and HPC workloads.