80GB VRAM GPU Cloud Rental for AI & LLMs

5 GPU models with approximately 80GB VRAM, prices from $0.452/hr. Best for: 70B model inference, fine-tuning large models.

Last updated May 26, 2026 · Data refreshed every 6 hours
GPU Models
5
Cheapest
$0.452/hr
Best $/GB
$0.006/GB-hr
Total Instances
632

Will this VRAM tier run my model?

80GB VRAM is best for 70B model inference, fine-tuning large models. For LLM inference, leave headroom for quantization format, batch size, and KV cache: long context windows can add meaningful VRAM pressure even when the base model fits.

All 80GB VRAM GPUs

GPU Model Providers Instances From / GB From
A100 10 211 $0.006/GB-hr $0.452/hr
H100 36 327 $0.010/GB-hr $0.801/hr
A100 80GB 30 72 $0.014/GB-hr $1.08/hr
A100 SXM 1 1 $0.020/GB-hr $1.60/hr
H100 SXM 13 21 $0.029/GB-hr $2.29/hr
Other VRAM Tiers