Report: DeepSeek is helping design Huawei's AI chips to train frontier models without Nvidia
kimmonismus · x · 2026-10-20
According to The Information, DeepSeek is working closely with Huawei on the design of future AI chips, aiming to make frontier-model training possible without Nvidia.
- Huawei engineers have been stationed at DeepSeek's offices, incorporating DeepSeek's requests into design decisions like how models compute and move data between processors — far beyond typical supplier roadmap discussions.
- The work covers chips shipping next year and beyond; TileLang software is also easing platform switching.
- Chip performance, manufacturing capacity, and large-cluster reliability remain major hurdles for China's full-stack independence.
Related event: DeepSeek Reportedly Co-Designing Huawei's Next-Gen AI Chips(2 posts)→
More from Infra
- 26-author study: a hardwired frontier AI training pause is technically feasible — wfithian · 2026-10-20
- Stardock CEO: local AI will dominate most AI work in 18 months, datacenters pointless — draginol · 2026-10-20
- Students get 1 year of Google AI Pro free, with 35 hours of monthly A100 on Colab — CeFurkan · 2026-10-20
- 768p video with audio generated on just two desktop DGX Sparks using new FastVideo QAD-v2 — Scobleizer · 2026-10-20
- AI power shortage is overstated: upgrades and exports shrink the gap, glut by 2028 — toptickcrypto · 2026-10-20
- Nebius launches live GPU spot pricing tracker for H100/H200/B200/B300 — demian_ai · 2026-10-20