LeituraPesquisaRadarFramework de investimento
Entrar / Cadastrar
← Voltar para a leitura
NotíciasTópico · 华为英伟达算力差距Material2 min de leituraEditorial DisconfirmAI

SemiAnalysis Test: Huawei Ascend Hardware Lags A100 by ~5 Years, DeepSeek Software Near Frontier

SemiAnalysis reports that while Huawei's Ascend 950PR hardware significantly trails NVIDIA's A100, DeepSeek's software optimizations have reached frontier levels, rapidly narrowing the AI chip ecosystem gap through mutual complementarity.

2026-10-23 15:57 GMT+8

In a report published on October 21, semiconductor research firm SemiAnalysis conducted independent benchmark tests on Huawei's Ascend 950PR chip and analyzed the current state of China's AI software-hardware synergy in light of DeepSeek's recent open-source toolchain updates. The core conclusion is that compared to the NVIDIA A100 baseline, Huawei's hardware performance may lag by more than five years, yet the software optimization tools provided by DeepSeek have already reached the frontier.

SemiAnalysis tested the Ascend 950PR, equipped with 28 cube compute cores and 128 GiB of High Bandwidth Memory (HBM), across 1,055 BF16 matrix multiplication (GEMM) operation shapes. Results showed that Huawei's best-performing kernel achieved up to 99.7% of its theoretical upper limit. This performance was realized using PyTorch's standard torch.matmul function executed by Huawei's own operator library, not DeepSeek's code. This indicates that DeepSeek's extreme optimization in DeepGEMM is not the only way to approach the upper limit.

However, in terms of absolute performance, the Ascend 950PR remains significantly behind. In memory-bandwidth-bound operations, the median ratio of actual achieved performance to theoretically achievable performance for Ascend was 41.5%, compared to 46.8% for the NVIDIA H100. When comparing absolute values, the 950PR falls roughly between NVIDIA's A100, launched in 2020, and AMD's MI300X. This serves as the basis for the report's claim that Huawei's hardware 'lags by more than five years.'

The report highlights a complementary relationship between Huawei and DeepSeek: Huawei possesses a mature hardware ecosystem but lacks a frontier lab model capable of proving its chips can be used for training or inference; DeepSeek has advanced models but faces constraints in accessing compute power. On September 30, DeepSeek released Ascend support for six repositories, including DeepGEMM and FlashMLA. However, at the time of writing, only one of these six tools—the third-party framework TileLang—was actually usable. The other five depended on Huawei's CANN 9.2.0 software suite, which had not yet been officially released, along with specific compiler support.

Entrar / Cadastrar
LeituraPesquisaRadarFramework de investimento

SemiAnalysis estimates that by 2027 or 2028, DeepSeek will likely primarily switch to Huawei's technology stack, though this is a prediction by the research firm rather than an announced plan by DeepSeek. The report also emphasizes that while this collaboration works for DeepSeek, it does not necessarily make Ascend equally user-friendly for other developers, noting that compatibility issues on x86 hosts remain unresolved.

Fontes

  • techorange.comTechOrange 科技報橘↗
Tópico

华为英伟达算力差距

【背景】美国对华半导体出口管制持续收紧,中国本土AI芯片龙头华为与全球霸主英伟达之间的算力差距成为衡量中美科技竞争的关键指标。尽管华为在软件生态上取得突破,但硬件性能与产能的双重瓶颈使其难以在2030年前追平英伟达。

6 etapas2 peçasAtualizado 2026-10-25 23:41

Como chegamos aqui

  1. 2025-12-15美国外交关系委员会(CFR)发布报告称,华为AI芯片性能落后英伟达5倍且差距扩大,建议维持出口管制。cfr.org ↗
  2. 2026-09-04Epoch AI研究员Venkat Somala和Lynette Bye发文,量化华为2026年总算力仅为英伟达1/25,预计2030年仍落后4年。x.com ↗
  3. 2026-09-24Epoch AI发布详细报告《How far behind Nvidia is Huawei?》,论证出口管制限制了华为芯片性能和产量两个扩展杠杆。epochai.substack.com ↗

O que ainda está indefinido

  • 华为CANN 9.2.0软件套件发布后,除TileLang外的其他5个DeepSeek工具能否在昇腾芯片上稳定运行?
  • 在非DeepSeek特定超节点架构(如标准x86主机)上,华为昇腾芯片的兼容性问题何时能解决?
  • DeepSeek是否在2027-2028年正式将主要训练和推理负载迁移至华为技术栈?

Também nesta thread

  • Epoch AI Analysis: Why Huawei Is Unlikely to Catch Up with Nvidia by 20302026-10-12
Ler o tópico completo →
Continuar pesquisando →