vast.ai

Explore real product flows

Your next GPU workload, powered by curiosity.

FLUX.1 + ComfyUI on RTX 4090 for $0.33/hr. 1-click deploy, unlimited generations. No API f

A100 80GB from $0.86/hr. Still the workhorse for fine-tuning. 33% cheaper than Lambda. No

A100 80GB from $0.86/hr. Still the workhorse. 33% cheaper than Lambda.

Dia 1.6B TTS on RTX 4090 for $0.33/hr. Multi-speaker, voice cloning, Apache 2.0. No per-mi

RTX 5090 from $0.42/hr. 697 available right now. 32GB GDDR7.

H200 from $2.19/hr. 80GB HBM3e, 65% cheaper than hyperscalers.

H100 SXM from $1.67/hr. 80GB HBM3, NVLink. Half the cost of reserved.

X ad