Flopper.io Flopper.io
Huawei logo vs

Huawei Ascend 960 vs NVIDIA GH200 144GB HBM3e

The Ascend 960 and the GH200 deliver near-identical FP8 throughput (2,000 vs 1,979 TFLOPS dense). The Ascend 960 also carries 144 GB more memory (288 GB vs 144 GB).

Ascend 960: Da Vinci v4 (chiplet), 2027 GH200: Hopper, 2024
Memory
288 vs 144 GB
144 GB more for the Ascend 960
Bandwidth
9.6 TB/s vs 4.9 TB/s
Ascend 960 moves data faster
Download comparison sheet

Share this page

Huawei Ascend 960 vs NVIDIA GH200 144GB HBM3e

A post with this link shows the card on the Image tab as its preview.

Post on X Share on LinkedIn
 Printable version →
Huawei logo

Huawei Ascend 960

Full specs →

What is the Huawei Ascend 960?

The Ascend 960 is an NPU from Huawei on the Da Vinci v4 (chiplet) architecture, an accelerator built for AI inference. It is an OAM module, the open accelerator format used in multi-GPU servers. It has been announced but is not yet shipping. Ascend 960 full specifications →

Can I rent the Ascend 960 in the cloud?

Flopper is not tracking any live rental listings for the Ascend 960 yet. The provider directory lists the clouds that rent comparable accelerators on demand. GPU cloud providers →

How does the Ascend 960 compare with other accelerators?

This page sets the Ascend 960 against the GH200. The Ascend 960 is the newer generation of the two. This is a cross-vendor matchup, Huawei against NVIDIA; the Ascend 960 is an OAM module while the GH200 is a superchip. The compare tool builds the same comparison for any two parts in the catalogue. All GPU comparisons →

Where do the Ascend 960 figures come from?

Huawei published the Ascend 960 figures in marketing material rather than a datasheet, and Flopper has no independent confirmation of them. How Flopper verifies specifications →

NVIDIA GH200 144GB HBM3e

Full specs →

What is the NVIDIA GH200?

The GH200 is a GPU from NVIDIA on the Hopper architecture, an accelerator for AI training, inference and high-performance computing. It is a superchip that pairs a CPU and accelerators on one board for rack-scale systems. GH200 full specifications →

Can I rent the GH200 in the cloud?

Flopper is not tracking any live rental listings for the GH200 yet. The provider directory lists the clouds that rent comparable accelerators on demand. GPU cloud providers →

How does the GH200 compare with other accelerators?

This page sets the GH200 against the Ascend 960. The GH200 is the older generation of the two. This is a cross-vendor matchup, NVIDIA against Huawei; the GH200 is a superchip while the Ascend 960 is an OAM module. The compare tool builds the same comparison for any two parts in the catalogue. All GPU comparisons →

Where do the GH200 figures come from?

Every GH200 figure on this page traces to NVIDIA's own datasheet or specification table. How Flopper verifies specifications →

Specifications

Huawei logo
Ascend 960
GH200
Architecture
Da Vinci v4 (chiplet)Hopper
Launch Year
20272024
Form Factor
OAMSuperchip
Memory
288 GB144 GB
Memory Bandwidth
9.6 TB/s4.9 TB/s
TDP
—1.0 kW
Process Node
—4nm

Performance (TFLOPS)

Huawei logo
Ascend 960
GH200
No verified data available34 TFLOPS
No verified data available67 TFLOPS
No verified data available494 TFLOPS
989 TFLOPS sparse
1,000 TFLOPS 990 TFLOPS
1,979 TFLOPS sparse
1,000 TFLOPS 990 TFLOPS
1,979 TFLOPS sparse
2,000 TFLOPS 1,979 TFLOPS
3,958 TFLOPS sparse
FP6
No verified data available No verified data available
4,000 TFLOPS No verified data available
No verified data available1,979 TOPS
3,958 TOPS sparse
Tensor Core
No verified data available67 TFLOPS

Figures are dense. Where a vendor publishes a figure it sits under the dense one and never decides a winner. Hover any precision for its definition, or read the GPU glossary.

FLOPS by Precision

What actually differs

The Ascend 960 is the newer part: Da Vinci v4 (chiplet) (2027) against the GH200's Hopper (2024).

The Ascend 960 carries 144 GB more memory (288 GB against 144 GB), so larger models fit on one GPU before you shard. Bandwidth is 9.6 TB/s against 4.9 TB/s, which sets the ceiling on decode speed.

Weight the FP8 and FP16 rows for LLM work and the FP64 row for HPC.

New to the terms? The glossary explains FP64, FP32, TF32, BF16, FP16, FP8, FP4, INT8, Sparsity and TFLOPS. For how the generations relate, see NVIDIA architectures explained.

Frequently asked questions

Is the Ascend 960 faster than the GH200?

They are close on paper: both deliver about 2,000 TFLOPS of dense FP8 throughput. Memory, bandwidth, and power are the deciding differences.

Which should you choose, the Ascend 960 or the GH200?

On raw FP8 throughput the Ascend 960 and the GH200 are level, so the decision rests on memory, power and price. Choose the Ascend 960 when the model has to fit on one GPU: it carries 144 GB more memory. Packaging differs too: the Ascend 960 is an OAM module while the GH200 is a superchip, so check what your servers accept.

Which has more memory, the Ascend 960 or the GH200?

The Ascend 960 carries 288 GB of memory versus 144 GB for the GH200. Memory bandwidth is 9.6 TB/s for the Ascend 960 and 4.9 TB/s for the GH200.

Get Comparison Updates

New GPUs added weekly. Be the first to see how they compare.

Open in Full Compare Tool → Ascend 960 full specs GH200 full specs Estimate cluster cost in the calculator →

© 2026 Flopper.io - Compare the GPUs Powering AI