search
CUDA
Trends
- 1
Chinese AI firm DeepSeek and Huawei have opened up programming tools for Huawei's Ascend AI chips, making the technology accessible to outside developers. The move is seen as a step toward building a domestic AI software ecosystem around Chinese hardware, reducing reliance on Nvidia's CUDA stack amid ongoing US export restrictions. Details on the exact terms of the release remain limited.
- 2
Chinese AI startup DeepSeek has introduced software that allows its AI models to run on Huawei's Ascend chips, positioning the pairing as a domestic alternative to Nvidia's hardware. The move underscores China's push for technological self-sufficiency amid US export restrictions on advanced chips, and intensifies the competition between Huawei's ecosystem and Nvidia's dominant CUDA software platform.
- 3DeepSeek Open-Sources Ascend Versions of Key AI Libraries▼DeepSeek Open-Sources Ascend Versions of TileLang, DeepGEMM and DeepEP as Huawei Details SuperPoD Flex
DeepSeek has released open-source Ascend-compatible versions of its TileLang, DeepGEMM and DeepEP libraries, extending its AI software stack to Huawei's chip platform. The release came alongside Huawei detailing SuperPoD Flex, its flexible large-scale computing cluster architecture. The moves highlight deepening cooperation between China's leading AI lab and Huawei as both work to build alternatives to Nvidia's CUDA ecosystem.
- 4DeepSeek Open-Sources Full Ascend Infrastructure Stack▼DeepSeek Open-Sources Full Ascend Infrastructure Stack, Achieving "One-to-One Parity" with Nvidia Platform
Chinese AI firm DeepSeek has open-sourced its full infrastructure stack built on Huawei's Ascend chips, claiming one-to-one parity with its Nvidia-based platform. The move gives developers outside the CUDA ecosystem a complete, openly available software stack for training and running large AI models on domestic hardware. It is being read as a significant step toward viable alternatives to Nvidia amid ongoing US export restrictions on advanced chips to China.
- 5DeepSeek and Huawei Unveil Open-Source Toolkit to Challenge Nvidia CUDA▼DeepSeek and Huawei Unveil Open-Source Toolkit to Loosen Nvidia's CUDA Grip
DeepSeek and Huawei have jointly released an open-source toolkit aimed at reducing reliance on Nvidia's CUDA software ecosystem. The move pairs DeepSeek's AI research with Huawei's hardware push, offering developers an alternative stack for training and running AI models on Chinese chips. Industry watchers see it as a significant step in China's effort to build a self-sufficient AI computing stack amid US export restrictions on advanced Nvidia hardware.
- 6DeepSeek open-sources toolkit for Huawei AI chips●DeepSeek has released an open-source software toolkit for Huawei's Ascend AI accelerators, including a programming langu
DeepSeek has released an open-source software toolkit for Huawei's Ascend AI accelerators, including a programming language called TileLang that rivals Nvidia's CUDA. The move supports Chinese AI development on domestic hardware and reduces reliance on Nvidia, whose chip sales to China face US export restrictions. Observers see it as a step toward an alternative software ecosystem for AI computing built around Chinese-made chips.
- 7DeepSeek Teams Up With Huawei to Challenge Nvidia's Software Lead●DeepSeek Targets Nvidia's Software Moat With Huawei Partnership
DeepSeek is working with Huawei in an effort to break into Nvidia's dominance of AI software, where Nvidia's CUDA ecosystem has long been a key competitive barrier. The partnership signals a push to build a viable Chinese alternative spanning hardware and software for AI computing, intensifying the technology rivalry between US and Chinese chipmakers.
Repos
- lostmsu/TurboGPT Train a tiny GPT in under a minute (CUDA only)
- tensorflow/tensorflow An Open Source Machine Learning Framework for Everyone
- magnitudedev/magnitude Open source inference engine for agents that optimizes itself for your exact hardware. Compiles and tunes its kernels on
- tile-ai/tilelang Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels