Huawei Ascend 910C vs Huawei Atlas 350

The Ascend 910C delivers 88% more FP16 throughput than the Atlas 350 (800 vs 425 TFLOPS dense). The Ascend 910C also carries 16 GB more memory (128 GB vs 112 GB).

Ascend 910C: Da Vinci v2 (dual-die), 2025 Atlas 350: Da Vinci v3, 2026

FP16 dense lead

+88%

Ascend 910C: 800 vs 425 TFLOPS

Memory

128 vs 112 GB

16 GB more for the Ascend 910C

Bandwidth

3.2 TB/s vs 1.4 TB/s

Ascend 910C moves data faster

TDP

600 W vs 600 W

Same power budget

Download comparison sheet Printable version →

Huawei Ascend 910C

Leads on FP16 Full specs →

What is the Huawei Ascend 910C?

The Ascend 910C is an NPU from Huawei on the Da Vinci v2 (dual-die) architecture, an accelerator built for AI inference. It is an OAM module, the open accelerator format used in multi-GPU servers. Ascend 910C full specifications →

Can I rent the Ascend 910C in the cloud?

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

How does the Ascend 910C compare with other accelerators?

This page sets the Ascend 910C against the Atlas 350. The Ascend 910C is the older generation of the two. Both are Huawei parts. The compare tool builds the same comparison for any two parts in the catalogue. All GPU comparisons →

Where do the Ascend 910C figures come from?

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

Huawei Atlas 350

Full specs →

What is the Huawei Atlas 350?

The Atlas 350 is an NPU from Huawei on the Da Vinci v3 architecture, an accelerator built for AI inference. Atlas 350 full specifications →

Can I rent the Atlas 350 in the cloud?

Flopper is not tracking any live rental listings for the Atlas 350 yet. The provider directory lists the clouds that rent comparable accelerators on demand. GPU cloud providers →

How does the Atlas 350 compare with other accelerators?

This page sets the Atlas 350 against the Ascend 910C. The Atlas 350 is the newer generation of the two. Both are Huawei parts. The compare tool builds the same comparison for any two parts in the catalogue. All GPU comparisons →

Where do the Atlas 350 figures come from?

Every Atlas 350 figure on this page traces to Huawei's own datasheet or specification table. How Flopper verifies specifications →

Specifications

Ascend 910C

Atlas 350

Architecture

Da Vinci v2 (dual-die)Da Vinci v3

Launch Year

20252026

Form Factor

OAMCustom

Memory

128 GB112 GB

Memory Bandwidth

3.2 TB/s1.4 TB/s

TDP

600 W600 W

Process Node

SMIC 7nm—

Spec

Ascend 910C

Atlas 350

Architecture Da Vinci v2 (dual-die)Da Vinci v3

Launch Year 20252026

Form Factor OAMCustom

Memory 128 GB112 GB

Memory Bandwidth 3.2 TB/s1.4 TB/s

TDP 600 W600 W

Process Node SMIC 7nm—

Performance (TFLOPS)

Ascend 910C

Atlas 350

50 TFLOPS

800 TFLOPS 425 TFLOPS

804 TFLOPS

FP6

804 TFLOPS

1,561 TFLOPS

1,600 TOPS 804 TOPS

Precision

Ascend 910C

Atlas 350

50 TFLOPS

800 TFLOPS 425 TFLOPS

804 TFLOPS

FP6 804 TFLOPS

1,561 TFLOPS

1,600 TOPS 804 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 350 is the newer part: Da Vinci v3 (2026) against the Ascend 910C's Da Vinci v2 (dual-die) (2025).

The Ascend 910C carries 16 GB more memory (128 GB against 112 GB), so larger models fit on one GPU before you shard. Bandwidth is 3.2 TB/s against 1.4 TB/s, which sets the ceiling on decode speed.

Power is 600 W against 600 W, and at FP32 the Ascend 910C 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, BF16, FP16, FP8, FP4, INT8, Sparsity and TFLOPS.

Frequently asked questions

What is better, the Ascend 910C or the Atlas 350?

The Ascend 910C, on throughput. At FP16 precision the Ascend 910C reaches 800 TFLOPS dense against 425 TFLOPS for the Atlas 350. The performance table on this page lists every published precision for both GPUs.

Which should you choose, the Ascend 910C or the Atlas 350?

Choose the Ascend 910C for training and high-throughput inference: it has 88% more FP16 throughput than the Atlas 350. It also carries 16 GB more memory, so larger models fit on one GPU before you shard.

Which has more memory, the Ascend 910C or the Atlas 350?

The Ascend 910C carries 128 GB of memory versus 112 GB for the Atlas 350. Memory bandwidth is 3.2 TB/s for the Ascend 910C and 1.4 TB/s for the Atlas 350.

How much power do the Ascend 910C and the Atlas 350 draw?

The Ascend 910C is rated at 600 W TDP and the Atlas 350 at 600 W. On FP32 throughput per watt, the Ascend 910C 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 910C full specs Atlas 350 full specs Estimate cluster cost in the calculator →

Huawei Ascend 910C vs Huawei Atlas 350

Huawei Ascend 910C

What is the Huawei Ascend 910C?

Can I rent the Ascend 910C in the cloud?

How does the Ascend 910C compare with other accelerators?

Where do the Ascend 910C figures come from?

Huawei Atlas 350

What is the Huawei Atlas 350?

Can I rent the Atlas 350 in the cloud?

How does the Atlas 350 compare with other accelerators?

Where do the Atlas 350 figures come from?

Specifications

Performance (TFLOPS)

FLOPS by Precision

What actually differs

Frequently asked questions

What is better, the Ascend 910C or the Atlas 350?

Which should you choose, the Ascend 910C or the Atlas 350?

Which has more memory, the Ascend 910C or the Atlas 350?

How much power do the Ascend 910C and the Atlas 350 draw?

Get Comparison Updates

Stay Updated on GPU Specs