Ascend 910C
Ascend 910C is a Huawei custom AI chip (ASIC) based on the Da Vinci architecture, first released in 2024.
Built from BenchGecko data as of April 14, 2026 · updates when the data changes
Ascend 910C is a Huawei custom AI chip (ASIC) based on the Da Vinci architecture, first released in 2024.
Basic
The Huawei Ascend 910C is a custom AI chip (ASIC) made by Huawei. Peak throughput is 800 TFLOPS at FP16/BF16. It carries 128 GB of HBM2e with 3.2 TB/s of memory bandwidth. Rated power (TDP) is 550 W. It is built on N7 · SMIC.
Deep
In the BenchGecko dataset it sits in the frontier tier: current-generation flagship silicon shipping at scale to hyperscalers. Disclosed buyers include ByteDance and Baidu. Specs come from manufacturer datasheets; the spec card on this page shows the dataset values and their as-of date.
Expert
That is 1.45 FP16 TFLOPS per watt. Memory bandwidth per unit of compute is 4 GB/s per FP16 TFLOP: a higher ratio helps memory-bound inference, while peak TFLOPS matter more for compute-bound training. Fabrication: China. Real throughput depends on the model, precision, batch size and software stack, so compare tokens per second per dollar for your own workload.
Depending on why you're here
- ·Ascend 910C: 800 TFLOPS FP16/BF16
- ·Process: N7 · SMIC
- ·Tier: frontier
- ·128 GB of HBM2e per chip
- ·No public cloud listing for Ascend 910C in the dataset
- ·Compare it with other chips on /hardware/ascend-910c
- ·Disclosed buyers: ByteDance and Baidu
- ·Huawei · fabbed at SMIC
- ·Supply tightness is tracked on /hardware
- ·Ascend 910C is a chip built to run AI, made by Huawei
- ·It has 128 GB of fast memory so large models fit
- ·Companies buy or rent these to train and serve AI models
Frequently Asked Questions
Read the primary sources
- Ascend overviewe.huawei.com