Flopper.io GPU Specification Sheet
October 9, 2026
flopper.io/gpu/aws-inferentia2AWS Inferentia2
NeuronCore-v2 2022
Overview
47.5
FP32 TFLOPS
32 GB
Memory HBM
820 GB/s
Memory Bandwidth
—
TDP
Performance Metrics
| Precision | Peak (dense) | Bits | Type |
|---|---|---|---|
| FP64 | — | — | Standard |
| FP32 | 47.5 | 32 | Standard |
| TF32 | 190.0 | 32 | Standard |
| BF16 | 190.0 | 16 | Standard |
| FP16 | 190.0 | 16 | Standard |
| FP8 | 190.0 | 8 | Standard |
| FP6 | — | — | Standard |
| FP4 | — | — | Standard |
| INT8 | 380.0 | 8 | Standard |
Memory Specifications
- Capacity
- 32 GB
- Type
- HBM
- Bandwidth
- 820 GB/s
- Interface Width
- —
Power & Efficiency
- TDP
- —
- Max power
- —
- FP32 dense TFLOPS per W of TDP
- —
- FP16 dense TFLOPS per W of TDP
- —
Per W of TDP figures divide dense peak by accelerator TDP. Board power only: excludes host CPUs, networking, cooling and facility overhead.
Hardware Details
Vendor: AWS
Architecture: NeuronCore-v2
Process Node: —
Form Factor: —
Launch Year: 2022
Compute Engine
NeuronCore-v2: 2
Interconnect
GPU-to-GPU: NeuronLink-v2
Bandwidth: 192 GB/s direction and scope not stated by vendor