Skip to main content

NVIDIA H20 (2024) vs Huawei Ascend 910B: Spec Comparison & Buyer's Guide

In AI infrastructure selection, NVIDIA H20 (2024) and Huawei Ascend 910B are two accelerators frequently compared. This article contrasts them item by item — architecture, compute, memory, power, and release cadence — to help you quickly judge which fits training or inference workloads.

Spec Comparison Table​

VendorNVIDIA H20 (2024)Huawei Ascend 910B
VendorNVIDIAHuawei
ArchitectureHopperDa Vinci
ProcessTSMC 4N (4nm)TSMC 7nm+
Release Date20242022-11-08
FP8 Compute296 TFLOPS—
FP16 Compute148 TFLOPS256 TFLOPS
FP32 Compute24 TFLOPS51.2 TFLOPS
INT8 Compute296 TOPS512 TOPS
Memory Type——
Memory Capacity—64 GB HBM2e/ HBM3e
Memory Bandwidth4.0 TB/s600 GB/s/ 1,200 GB/s
TDP Power400W310 W

Key Differences​

  • Power: Huawei Ascend 910B has a TDP of 310 W, lower than NVIDIA H20 (2024)'s 400W, friendlier to datacenter PUE and cooling.

Selection Advice​

  • When chasing extreme single-card compute and a mature toolchain, prioritize NVIDIA H20 (2024); if budget, power wall, or local support are hard constraints, Huawei Ascend 910B often fits better. Use this site's AI Compute Card Comparison Tool to validate multiple chips side-by-side before deciding.

FAQ​

What are the main differences between NVIDIA H20 (2024) and Huawei Ascend 910B?​

The core difference is architecture and compute density: NVIDIA H20 (2024) uses Hopper, FP8 ~296 TFLOPS, memory —; Huawei Ascend 910B uses Da Vinci, FP8 ~No public FP8 data, memory 64 GB HBM2e/ HBM3e. See the comparison table above.

What is the TDP (power) of NVIDIA H20 (2024)?​

NVIDIA H20 (2024) has a TDP of 400W; actual whole-system power also includes board, fans, and PUE.

Which is better for large-model training / inference?​

Training values memory capacity, bandwidth, and multi-card interconnect; inference values single-card throughput and power efficiency. Combine the "Key Differences" and "Selection Advice" above with your batch size, model size, and SLA.

How much do NVIDIA H20 (2024) and Huawei Ascend 910B differ in memory capacity?​

NVIDIA H20 (2024) is —, Huawei Ascend 910B is 64 GB HBM2e/ HBM3e; the gap directly affects loadable model size and context length.