Flopper.io GPU Specification Sheet
September 26, 2026
flopper.io/gpu/huawei-atlas-350Huawei Atlas 350
Da Vinci v3 Custom 2026
Overview
—
FP32 TFLOPS
112 GB
Memory HiBL 1.0
1.4 TB/s
Memory Bandwidth
600 W
TDP
Performance Metrics
| Precision | Peak (dense) | Bits | Type |
|---|---|---|---|
| FP64 | — | — | Standard |
| FP32 | — | — | Standard |
| TF32 | — | — | Standard |
| BF16 | 425.0 | 16 | Standard |
| FP16 | 425.0 | 16 | Standard |
| FP8 | 804.0 | 8 | Standard |
| FP6 | 804.0 | — | Standard |
| FP4 | 1561.0 | 4 | Standard |
| INT8 | 804.0 | 8 | Standard |
Memory Specifications
- Capacity
- 112 GB
- Type
- HiBL 1.0
- Bandwidth
- 1.4 TB/s
- Interface Width
- —
Power & Efficiency
- TDP
- 600 W
- Max power (Flopper estimate)
- ~690 W est.
- FP32 dense TFLOPS per W of TDP
- —
- FP16 dense TFLOPS per W of TDP
- 0.71 TFLOPS/W
Flopper estimate: 600 W TDP x 1.15. The vendor publishes no maximum board power for this part. 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: Huawei
Architecture: Da Vinci v3
Process Node: —
Form Factor: Custom
Launch Year: 2026
Interconnect
GPU-to-GPU: Unified Bus 2.0