Flopper.io Flopper.io
Huawei logo

Huawei Ascend 910A vs Huawei Atlas 300I Duo 96GB

The Ascend 910A delivers 83% more FP16 throughput than the Atlas 300I Duo (256 vs 140 TFLOPS dense). The Atlas 300I Duo counters with 64 GB more memory (96 GB vs 32 GB).

Ascend 910A: Da Vinci, 2019 Atlas 300I Duo: Da Vinci, 2022
FP16 dense lead
+83%
Ascend 910A: 256 vs 140 TFLOPS
Memory
32 vs 96 GB
64 GB more for the Atlas 300I Duo
Bandwidth
1.2 TB/s vs 408 GB/s
Ascend 910A moves data faster
TDP
310 W vs 150 W
Atlas 300I Duo draws 160 W less
Download comparison sheet

Share this page

Huawei Ascend 910A vs Huawei Atlas 300I Duo 96GB

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 910A

Leads on FP16 Full specs →

What is the Huawei Ascend 910A?

The Ascend 910A is an NPU from Huawei on the Da Vinci architecture, an accelerator built for AI inference. It is an OAM module, the open accelerator format used in multi-GPU servers. Ascend 910A full specifications →

Can I rent the Ascend 910A in the cloud?

Flopper is not tracking any live rental listings for the Ascend 910A yet. The provider directory lists the clouds that rent comparable accelerators on demand. GPU cloud providers →

How does the Ascend 910A compare with other accelerators?

This page sets the Ascend 910A against the Atlas 300I Duo. The Ascend 910A is the older generation of the two. Both are Huawei parts; the Ascend 910A is an OAM module while the Atlas 300I Duo is a PCIe card. The compare tool builds the same comparison for any two parts in the catalogue. All GPU comparisons →

Where do the Ascend 910A figures come from?

Every Ascend 910A figure on this page traces to Huawei's own datasheet or specification table. How Flopper verifies specifications →

Huawei logo

Huawei Atlas 300I Duo 96GB

Full specs →

What is the Huawei Atlas 300I Duo?

The Atlas 300I Duo is an NPU from Huawei on the Da Vinci architecture, an accelerator built for AI inference. It is a PCIe add-in card, so it fits standard servers and workstations. Atlas 300I Duo full specifications →

Can I rent the Atlas 300I Duo in the cloud?

Flopper is not tracking any live rental listings for the Atlas 300I Duo yet. The provider directory lists the clouds that rent comparable accelerators on demand. GPU cloud providers →

How does the Atlas 300I Duo compare with other accelerators?

This page sets the Atlas 300I Duo against the Ascend 910A. The Atlas 300I Duo is the newer generation of the two. Both are Huawei parts; the Atlas 300I Duo is a PCIe card while the Ascend 910A is an OAM module. The compare tool builds the same comparison for any two parts in the catalogue. All GPU comparisons →

Where do the Atlas 300I Duo figures come from?

Every Atlas 300I Duo figure on this page traces to Huawei's own datasheet or specification table. How Flopper verifies specifications →

Specifications

Huawei logo
Ascend 910A
Huawei logo
Atlas 300I Duo
Architecture
Da VinciDa Vinci
Launch Year
20192022
Form Factor
OAMPCIe
Memory
32 GB96 GB
Memory Bandwidth
1.2 TB/s408 GB/s
TDP
310 W150 W
Process Node
TSMC 7nm—

Performance (TFLOPS)

Huawei logo
Ascend 910A
Huawei logo
Atlas 300I Duo
No verified data available No verified data available
16 TFLOPS No verified data available
No verified data available No verified data available
No verified data available No verified data available
256 TFLOPS 140 TFLOPS
No verified data available No verified data available
FP6
No verified data available No verified data available
No verified data available No verified data available
512 TOPS 280 TOPS

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 Atlas 300I Duo is the newer part: Da Vinci (2022) against the Ascend 910A's Da Vinci (2019).

The Atlas 300I Duo carries 64 GB more memory (96 GB against 32 GB), so larger models fit on one GPU before you shard. Bandwidth is 1.2 TB/s against 408 GB/s, which sets the ceiling on decode speed.

Power is 310 W against 150 W, and at FP32 the Ascend 910A does more work per watt. Weight the FP8 and FP16 rows for LLM work and the FP64 row for HPC.

New to the terms? The glossary explains FP32, FP16, INT8, Sparsity and TFLOPS.

Frequently asked questions

Is the Ascend 910A faster than the Atlas 300I Duo?

At FP16 precision the Ascend 910A reaches 256 TFLOPS dense against 140 TFLOPS for the Atlas 300I Duo. The performance table on this page lists every published precision for both GPUs.

Which should you choose, the Ascend 910A or the Atlas 300I Duo?

Choose the Ascend 910A for training and high-throughput inference: it has 83% more FP16 throughput than the Atlas 300I Duo. Choose the Atlas 300I Duo when the model has to fit on one GPU: it carries 64 GB more memory. The Ascend 910A draws 160 W more power, so budget for cooling and power delivery before you commit. Packaging differs too: the Ascend 910A is an OAM module while the Atlas 300I Duo is a PCIe card, so check what your servers accept.

Which has more memory, the Ascend 910A or the Atlas 300I Duo?

The Atlas 300I Duo carries 96 GB of memory versus 32 GB for the Ascend 910A. Memory bandwidth is 1.2 TB/s for the Ascend 910A and 408 GB/s for the Atlas 300I Duo.

How much power do the Ascend 910A and the Atlas 300I Duo draw?

The Ascend 910A is rated at 310 W TDP and the Atlas 300I Duo at 150 W. On FP32 throughput per watt, the Ascend 910A is the more efficient part.

Get Comparison Updates

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

Open in Full Compare Tool → Ascend 910A full specs Atlas 300I Duo full specs Estimate cluster cost in the calculator →

© 2026 Flopper.io - Compare the GPUs Powering AI