GoKawiilHuawei's impressive next-generation Ascend 900-series AI accelerators will be offered only in China, not internationally, as the company struggles to meet domestic demand amid capacity constraints, the company announced this week. While the upcoming Ascend 960-series neural processing units (NPUs) could rival some of AMD's and Nvidia's existing AI GPUs, demand for these units outside of China was not guaranteed anyway.
"Since we do not have enough capacity to even satisfy the demand in China, we do not have a plan to expand into the international market in a fully-fledged way," said Eric Xu, rotating chairman of Huawei, on the sidelines of the company's Huawei Connect conference, Reuters reports. He added that Huawei supplies limited volumes to 'some countries where demand is particularly strong,' though he did not elaborate.
Huawei this week unveiled its latest AI accelerator roadmap, revealing major training and inference performance gains for its next-generation Ascend 960, 970, and 980 NPUs over the existing Ascend 910C and Ascend 950-series. The Ascend 960DT and 960PR are set to increase their FP8 training performance to 2 PFLOPS and their FP4 inference performance to 4 PFLOPS and 8 PFLOPS, respectively, in 2027. Meanwhile, their successors, Ascend 970 and Ascend 980, are projected to increase their FP4 performance to 14 PFLOPS and 28 PFLOPS, respectively, in the coming years.
Latest Videos From Tom's Hardware Watch full video here:
Huawei Ascend vs Nvidia AI GPUs
Swipe to scroll horizontally NPU FP8 Performance FP4 Perf Memory Memory Bandwidth Interconnect Bandwidth Targeted Release Nvidia H200 4 PFLOPS - 141 GB HBM3E 4.8 TB/s 900 GB/s 2023 Q4 Nvidia B300 10 PFLOPS 15/20 S/D PFLOPS 279 GB HBM3E 8 TB/s 1.8 TB/s 2025 Q4 Ascend 950PR 1 PFLOPS 2 PFLOPS 128 GB of HiBL 1.0 1.6 TB/s 2 TB/s 2026 Q1 Ascend 950DT 1 PFLOPS 2 PFLOPS 144 GB of HiZQ 2.0 4.0 TB/s 2 TB/s 2026 Q4 Nvidia R200 17.5 PFLOPS 35/50 T/I PFLOPS 288 GB HBM4 19.2 TB/s 3 TB/s 2026 Q4 Ascend 960DT 2 PFLOPS 4 PFLOPS 288 GB 9.6 TB/s 2.2 TB/s 2027 Q1 Ascend 960PR 2 PFLOPS 8 PFLOPS 192 GB 2.4 TB/s 2.2 TB/s 2027 Q3 Ascend 970 3.6 PFLOPS 14 PFLOPS 288 GB 14.4 TB/s 4.4 TB/s 2028 Ascend 980 7.2 PFLOPS* 28 PFLOPS* 384 GB 38.4 TB/s* 8 TB/s 2029
*Preliminary data
S/D - Sparse and Dense
T/I - Training and Inference
But while the upcoming Ascend NPUs will be considerably faster than their predecessors, particularly for inference, they will remain well behind Nvidia's previous- and current-generation accelerators, at least in raw compute performance. Huawei's 2027 Ascend 960DT is projected to deliver 2 FP8 TFLOPS for training, compared with Nvidia's 4 FP8 TFLOPS for the H200, released in 2023. The Ascend 960PR is expected to offer 8 FP4 PFLOPS for training, which is far behind Nvidia's B300, which delivers 15–20 NVFP4 PFLOPS. Even the Ascend 980, targeted for 2029, is projected to reach 7.2 FP8 PFLOPS and 28 FP4 PFLOPS, well below Nvidia's R200, which is on track to deliver 17.5 FP8 PFLOPS and 35/50 FP4 PFLOPS this year.
... continue reading