GPU rental prices & what runs on each card ($0 markup)
Real hourly rates for 20 rentable GPUs (from $0.06/hr, as of 2026-07-09, $0 markup), what fits in each card's VRAM, and the honest own-vs-rent-vs-API math.
Every rentable GPU — cheapest real rate first
NVIDIA RTX 4060 8GB: from $0.08/hr (Vast.ai, as of 2026-07-09) · 238 of 565 catalog models fit its 8 GB.
NVIDIA RTX 3080 10GB: from $0.14/hr (Vast.ai, as of 2026-07-09) · 287 of 565 catalog models fit its 10 GB.
NVIDIA RTX 3060 12GB: from $0.06/hr (Vast.ai, as of 2026-07-09) · 312 of 565 catalog models fit its 12 GB.
NVIDIA RTX 4070 SUPER 12GB: from $0.10/hr (Vast.ai, as of 2026-07-09) · 312 of 565 catalog models fit its 12 GB.
NVIDIA RTX 4070 12GB: from $0.12/hr (Vast.ai, as of 2026-07-09) · 312 of 565 catalog models fit its 12 GB.
NVIDIA RTX 5070 12GB: from $0.22/hr (Vast.ai, as of 2026-07-09) · 312 of 565 catalog models fit its 12 GB.
NVIDIA RTX 5080 16GB: from $0.21/hr (Vast.ai, as of 2026-07-09) · 360 of 565 catalog models fit its 16 GB.
NVIDIA RTX 4080 SUPER 16GB: from $0.21/hr (Vast.ai, as of 2026-07-09) · 360 of 565 catalog models fit its 16 GB.
NVIDIA Tesla P40 24GB: from $0.13/hr (Vast.ai, as of 2026-07-09) · 402 of 565 catalog models fit its 24 GB.
NVIDIA RTX 3090 24GB: from $0.26/hr (Vast.ai, as of 2026-07-09) · 402 of 565 catalog models fit its 24 GB.
NVIDIA RTX 4090 24GB: from $0.37/hr (TensorDock, as of 2026-07-09) · 402 of 565 catalog models fit its 24 GB.
NVIDIA RTX 5090 32GB: from $0.56/hr (Vast.ai, as of 2026-07-09) · 429 of 565 catalog models fit its 32 GB.
NVIDIA RTX A5000 24GB: from $0.27/hr (RunPod, as of 2026-07-09) · 402 of 565 catalog models fit its 24 GB.
NVIDIA L4 24GB: from $0.39/hr (RunPod, as of 2026-07-09) · 402 of 565 catalog models fit its 24 GB.
NVIDIA A100 40GB: from $1.29/hr (Verda, as of 2026-07-09) · 446 of 565 catalog models fit its 40 GB.
NVIDIA RTX A6000 48GB: from $0.49/hr (RunPod, as of 2026-07-09) · 456 of 565 catalog models fit its 48 GB.
NVIDIA L40S 48GB: from $0.99/hr (RunPod, as of 2026-07-09) · 456 of 565 catalog models fit its 48 GB.
NVIDIA A100 80GB PCIe: from $1.39/hr (RunPod, as of 2026-07-09) · 487 of 565 catalog models fit its 80 GB.
NVIDIA H100 80GB SXM: from $2.25/hr (TensorDock, as of 2026-07-09) · 487 of 565 catalog models fit its 80 GB.
NVIDIA H200 141GB: from $3.44/hr (DigitalOcean, as of 2026-07-09) · 507 of 565 catalog models fit its 141 GB.
As an Amazon Associate we earn from qualifying purchases. Each link opens an Amazon search for the category — not an endorsement of any specific product — and Cynosure LLC may earn a commission at no extra cost to you.
A short email of real AI price moves, straight from the daily log — no hype. We're collecting the list now; the first issue goes out when it opens. Unsubscribe with one click.