Huawei unveils new Atlas 350 AI accelerator with 1.56 PFLOPS of FP4 compute and up to 112GB of HBM — claims 2.8x more performance than Nvidia's H20
Huawei unveiled the Atlas 350 AI accelerator at its China Partner Conference, featuring the Ascend 950PR chip, 1.56 PFLOPS FP4 performance, and 112GB of proprietary HiBL 1.0 HBM. The device targets the inference prefill stage and claims 2.8x the performance of Nvidia's H20. Bullish for Huawei's proprietary HiBL 1.0 HBM supply chain.
- Related manufacturers
- Huawei, Nvidia, SK Hynix, Micron
- Part numbers
- Atlas 350Ascend 950PRHiBL 1.0
- Impact score
- 25 / 100, low impact