2× RTX 4070 SUPER rentals from Vast.ai

1 provider with 12 offers. Sorted by most reliable. Updated 2026-11-12 06:17 UTC.

Vast.ai

marketplace12 offers

★★★★★★★★★★ 5.0

From host reliability5.0

From$0.081per GPU-hour · median $0.104

12 offers in stock · 40 GPUs free99.0% host reliabilityIn stock 100.0% of 30 days
  • 8× GPU · 12 GBJiangsu, CN
    8 GPUs free now
    $0.081per GPU-hour · about $119/mo
    View deal
  • 4× GPU · 12 GBJiangsu, CN
    4 GPUs free now
    $0.082per GPU-hour · about $120/mo
    View deal
  • 6× GPU · 12 GBAnhui, CN
    6 GPUs free now
    $0.095per GPU-hour · about $139/mo
    View deal
  • 2× GPU · 12 GBChina, CN
    2 GPUs free now
    $0.096per GPU-hour · about $140/mo
    View deal
  • 4× GPU · 12 GBAnhui, CN
    4 GPUs free now
    $0.096per GPU-hour · about $140/mo
    View deal
See all 12 offers from Vast.ai ›

Where to find RTX-4070-SUPER

Every provider we track, by region. 3 of 36 providers have it in stock now.

ProviderAmericas2 in stock · from $0.109Europe1 in stock · from $0.117Asia-Pacific1 in stock · from $0.081Elsewhere1 in stock · from $0.121
Vast.ai★★★★★★★★★★ 5.0From $0.1097 in stockNot offeredFrom $0.08114 in stockNot offered
Clore.ai★★★★★★★★★★ 4.0 (392)Not offeredFrom $0.1176 in stockNot offeredFrom $0.1211 in stock
SimplePodNot ratedFrom $0.2002 in stockNot offeredNot offeredNot offered

Not offered by: RunPod, Akash, Cudo Compute, io.net, Koyeb, Thunder Compute, Akamai Linode, Microsoft Azure, Oracle Cloud, AWS, OVHcloud, Lium, Exoscale, Nosana, GPU.ai, IMWT (via Shadeform), Massed Compute (via Shadeform), Lambda (via Shadeform), Crusoe (via Shadeform), Paperspace (via Shadeform), Hyperstack (via Shadeform), Boost Run (via Shadeform), Latitude.sh (via Shadeform), Scaleway (via Shadeform), Voltage Park (via Shadeform), Denvr (via Shadeform), Excess Supply (via Shadeform), Horizon (via Shadeform), Verda (via Shadeform), DigitalOcean (via Shadeform), Nebius (via Shadeform), Vultr, Phyntec (via Shadeform).

For agents

Free to use from code: a JSON API, an MCP server, and every page as Markdown. No API key.

curl "https://gpuscout.cc/api/v1/providers/ranking?gpu=h100&count=8"

llms.txt · OpenAPI · MCP endpoint: https://gpuscout.cc/mcp