GPU Catalog
Filter accelerators by VRAM, price, provider and availability — all resolved server-side.
112 accelerators with specs and real per-hour pricing across every provider that lists them.
- VM.GPU.A10.1 (1x A10) · 24 GB$1.05/hr
- 1x A10 24GB · 24 GB$0.75/hr
- g5.xlarge (1x A10G) · 24 GB$1.01/hr
- 1x A100 40GB · 40 GB$1.29/hr
- 8x A100 80GB · 640 GB$14.30/hr
- ND A100 v4 (8x A100 40GB) · 320 GB$19.68/hr
- BM.GPU.A100 v2 (8x 80GB) · 640 GB$16.00/hr
- A100 · 80 GB$0.64/hr
- NVIDIA A100 80GB · 80 GB$1.19/hr
- NVIDIA A100 80GB PCIe · 80 GB$1.59/hr
- A100 · 80 GB$1.30/hr
- A100 · 80 GB$2.30/hr
- A100 · 80 GB$3.40/hr
- a2-ultragpu-8g (8x A100 80GB) · 640 GB$18.58/hr
- A100 · 80 GB$7.52/hr
- p4d.24xlarge (8x A100 40GB) · 320 GB$19.38/hr
- A100 · 80 GB$21.60/hr
- A100 80GB SXM · 80 GB$1.49/hr
- A100 (80 GB SXM4) · 80 GB$22.32/hr
- A10G · 24 GB$0.45/hr
- A40 · 48 GB$0.44/hr
- A40 · 48 GB$0.35/hr
- A6000 · 48 GB$1.89/hr
- HGX B200 192GB · 192 GB$5.99/hr
- NVIDIA B200 192GB · 192 GB$5.99/hr
- B200 192GB · 192 GB$5.49/hr
- B200 (180 GB SXM6) · 180 GB$53.52/hr
- B300 · 288 GB$7.39/hr
- GB200 NVL72 (per-GPU) · 192 GB$6.99/hr
- GB200 NVL72 (per-GPU) · 192 GB$7.49/hr
- GB200 NVL72 · 186 GB$42.00/hr
- GH200 · 96 GB$6.50/hr
- GH200 (96 GB) · 96 GB$2.29/hr
- ND H100 v5 (8x H100) · 640 GB$30.60/hr
- NVIDIA H100 PCIe · 80 GB$2.49/hr
- NVIDIA H100 SXM5 · 80 GB$2.99/hr
- NVIDIA H100 SXM · 80 GB$1.99/hr
- p5.48xlarge (8x H100) · 640 GB$31.28/hr
- a3-highgpu-8g (8x H100) · 640 GB$29.86/hr
- HGX H100 80GB · 80 GB$2.29/hr
- 1x H100 SXM 80GB · 80 GB$2.99/hr
- 8x H100 SXM · 640 GB$23.92/hr
- HGX H100 (8x) · 640 GB$23.60/hr
- HGX H100 (8x) · 640 GB$23.12/hr
- BM.GPU.H100 (8x) · 640 GB$28.00/hr
- H100 · 80 GB$6.88/hr
- H100 · 80 GB$11.06/hr
- H100 · 80 GB$3.71/hr
- H100 · 80 GB$5.49/hr
- H100 · 80 GB$18.37/hr
- H100 · 80 GB$2.25/hr
- H100 · 80 GB$1.99/hr
- H100 · 80 GB$5.95/hr
- H100 (80 GB SXM5) · 80 GB$31.92/hr
- HGX H200 (8x) · 1128 GB$29.60/hr
- H200 · 141 GB$4.29/hr
- H200 · 141 GB$2.30/hr
- NVIDIA H200 SXM · 141 GB$2.89/hr
- ND H200 v5 (8x H200) · 1128 GB$41.20/hr
- p5en.48xlarge (8x H200) · 1128 GB$41.80/hr
- BM.GPU.H200 (8x) · 1128 GB$36.80/hr
- HGX H200 141GB · 141 GB$3.49/hr
- 8x H200 141GB · 1128 GB$31.20/hr
- HGX H200 (8x) · 1128 GB$30.40/hr
- NVIDIA H200 SXM5 141GB · 141 GB$3.99/hr
- a3-ultragpu-8g (8x H200) · 1128 GB$40.50/hr
- H200 NVL · 143 GB$3.79/hr
- HGX H100 · 80 GB$5.95/hr
- HGX H200 · 141 GB$50.44/hr
- HGX H200 · 200 GB$4.50/hr
- L4 · 24 GB$2.46/hr
- g2-standard-4 (1x L4) · 24 GB$0.71/hr
- g6.xlarge (1x L4) · 24 GB$0.80/hr
- L40 · 48 GB$0.95/hr
- NVIDIA L40S 48GB · 48 GB$1.39/hr
- g6e.xlarge (1x L40S) · 48 GB$1.86/hr
- BM.GPU.L40S.4 (per-GPU) · 48 GB$1.60/hr
- L40S · 48 GB$1.50/hr
- NVIDIA L40S 48GB · 48 GB$1.19/hr
- L40S · 48 GB$0.99/hr
- L40S · 48 GB$18.00/hr
- NVIDIA L40S 48GB · 48 GB$1.29/hr
- NVIDIA L40S 48GB · 48 GB$1.79/hr
- L40S · 48 GB$1.86/hr
- M4000 · 8 GB$0.45/hr
- MI300X · 192 GB$3.45/hr
- MI300X · 192 GB$2.39/hr
- P6000 · 24 GB$1.10/hr
- PRO 6000 MIG 48GB · 48 GB$1.00/hr
- RTX 4090 · 24 GB$0.47/hr
- NVIDIA RTX 4090 · 24 GB$0.59/hr
- NVIDIA RTX 4090 · 24 GB$0.79/hr
- RTX5000 · 16 GB$0.82/hr
- NVIDIA RTX 5090 · 32 GB$0.69/hr
- NVIDIA RTX 5090 · 32 GB$0.89/hr
- RTX 6000 (24 GB) · 24 GB$0.69/hr
- RTX A6000 · 48 GB$0.45/hr
- RTX PRO 6000 · 218 GB$1.80/hr
- RTX PRO 6000 BLACKWELL SERVER EDITION · 96 GB$20.00/hr
- RTX PRO 6000 WK · 96 GB$1.89/hr
- T4 · 16 GB$0.96/hr
- T4 · 16 GB$0.56/hr
- Tesla V100 · 16 GB$0.19/hr
- Tesla V100 (16 GB) · 16 GB$6.32/hr
- TESLA V100 SXM2 · 32 GB$0.17/hr
- TPU v4 (8-chip pod slice) · 256 GB$2.46/hr
- TPU v5e (8-chip pod slice) · 128 GB$1.52/hr
- TPU v6e Trillium (8-chip pod slice) · 256 GB$2.20/hr
- V100 · 16 GB$2.30/hr
- V100 · 32 GB$3.14/hr
- V100 · 16 GB$3.04/hr
- V100 SXM2 · 16 GB$0.23/hr