Accelerators pricing
AWS Inferentia prices in ap-northeast-1 (Asia Pacific (Tokyo))
Updated 2026-10-19
inf1.xlarge $0.308/h, inf1.2xlarge $0.489/h, inf1.6xlarge $1.59/h, inf1.24xlarge $6.38/h, inf2.xlarge $1.14/h, inf2.8xlarge $2.95/h, inf2.48xlarge $19.47/h.
Sizes
| Instance | Accelerators | Accel. memory | On-demand $/h | $/chip-hr | 3-yr RI no upfront $/h | Hours per $25k |
|---|---|---|---|---|---|---|
| inf1.xlarge | Inferentia ×1 | n/a | $0.308 | $0.308 | $0.148 | 81,169 h |
| inf1.2xlarge | Inferentia ×1 | n/a | $0.489 | $0.489 | $0.235 | 51,125 h |
| inf1.6xlarge | Inferentia ×4 | n/a | $1.59 | $0.3985 | $0.766 | 15,684 h |
| inf1.24xlarge | Inferentia ×16 | n/a | $6.38 | $0.3985 | $3.06 | 3,921 h |
| inf2.xlarge | Inferentia2 ×1 | 32 GB | $1.14 | $1.14 | $0.5241 | 21,982 h |
| inf2.8xlarge | Inferentia2 ×1 | 32 GB | $2.95 | $2.95 | $1.36 | 8,469 h |
| inf2.48xlarge | Inferentia2 ×12 | 384 GB | $19.47 | $1.62 | $8.97 | 1,284 h |
GPU instances in ap-northeast-1 for comparison
| Instance | Accelerators | Accel. memory | On-demand $/h | $/chip-hr | 3-yr RI no upfront $/h | Hours per $25k |
|---|---|---|---|---|---|---|
| g4dn.xlarge | NVIDIA T4 Tensor Core ×1 | 16 GB | $0.71 | $0.71 | $0.36 | 35,211 h |
| g5.xlarge | NVIDIA A10G ×1 | 24 GB | $1.46 | $1.46 | $0.6303 | 17,135 h |
| g6.xlarge | NVIDIA L4 ×1 | 24 GB | $1.17 | $1.17 | $0.5357 | 21,419 h |
| p4d.24xlarge | NVIDIA A100 ×8 | 320 GB | $30.10 | $3.76 | $15.79 | 831 h |
| p5.48xlarge | NVIDIA H100 ×8 | 640 GB | $68.80 | $8.60 | $29.72 | 363 h |
Other regions
Related cost calculators
- EC2 Pricing Calculator · Compare Instance Types & Costs
- Amazon SageMaker AI Pricing Calculator · On-Demand Studio, Training & Inference
Prices and credit amounts are estimates and may change. Check AWS for current terms.