Huawei Ascend 910A vs Huawei Atlas 200I A2 20 TOPS
The Ascend 910A delivers 25.6x the FP16 throughput of the Atlas 200I A2 (256 vs 10 TFLOPS dense). The Ascend 910A also carries 20 GB more memory (32 GB vs 12 GB).
Huawei Ascend 910A
Full specs →Huawei Atlas 200I A2 20 TOPS
Full specs →Specifications
Performance (TFLOPS)
FLOPS by Precision
What actually differs
The Atlas 200I A2 is the newer part: Da Vinci, launched in 2023, against the Ascend 910A's Da Vinci from 2019. Newer architectures typically add lower-precision formats and better throughput per watt, so check the precision rows your workload actually uses.
Dense throughput for the Ascend 910A against the Atlas 200I A2: FP16 256 vs 10 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.
Memory is 32 GB against 12 GB, fed at 1.2 TB/s versus 51 GB/s. More memory per GPU means larger models fit before you have to shard across cards, which often matters more than raw TFLOPS for inference.
For LLM training and inference, weight the FP8 and FP16 rows and memory capacity most heavily. For scientific and HPC workloads, the FP64 row is the one to read.
Frequently asked questions
Is the Ascend 910A faster than the Atlas 200I A2?
At FP16 precision the Ascend 910A reaches 256 TFLOPS dense against 10 TFLOPS for the Atlas 200I A2. The performance table on this page lists every published precision for both GPUs.
Which has more memory, the Ascend 910A or the Atlas 200I A2?
The Ascend 910A carries 32 GB of memory versus 12 GB for the Atlas 200I A2. Memory bandwidth is 1.2 TB/s for the Ascend 910A and 51 GB/s for the Atlas 200I A2.
How much power do the Ascend 910A and the Atlas 200I A2 draw?
The Ascend 910A is rated at 310 W TDP and the Atlas 200I A2 at 25 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.