Vast.ai
GPU marketplace — rent consumer cards at half the hyperscaler price.
Paid☁ Cloud · no GPU
Watermark-FreeHobbyist-FriendlyAPI
RunPod
On-demand GPU pods for ComfyUI, vLLM, training.
Paid 8–80 GB VRAMSelf-hosted server
Watermark-FreeHobbyist-FriendlyAPI
Visit RunPodUpdated 2026-05-16 · Affiliate link
Hardware requirements
Self-hosted server · Datacenter GPU (80 GB+)
Min VRAM
8 GB
Rec. VRAM
80 GB
Min RAM
16 GB
Rec. RAM
128 GB
Disk
100 GB
GPU class
Datacenter GPU
Whatever the pod image bundlesNo Apple SiliconGPU RequiredQuant: FP16, BF16, FP8 +1
Card mix ranges from RTX 3090 to H200 / B200.
Screenshot placeholder · RunPod
What is RunPod?
Rent A100 / H100 / 4090 / L40S pods by the minute. Templates for ComfyUI, A1111, vLLM, Ollama. Useful when your workflow outgrows your local box.
Pros & cons
Pros
- Cheaper than the big clouds
- Per-minute billing
- Ready-made templates
Cons
- Spot pods can be reclaimed
- You manage your own storage
- Network egress costs
What's actually free?
Pay-as-you-go. No free tier. Community-cloud pricing starts around $0.20/hr for older cards.
Watermark-Free
