DeepSeek has released six open-source software tools designed to help Huawei's Ascend AI accelerators compete with Nvidia's hardware, marking a significant step in China's effort to build an independent AI computing ecosystem. Announced on September 30, 2026, via DeepSeek's official channels, the toolkit includes TileLang—a high-level programming language positioned as a simpler alternative to Nvidia's CUDA—alongside five complementary libraries: DeepGEMM, FlashMLA, TileKernel, DeepSelect, and DeepEP. The software was developed in partnership with Huawei and optimized for the Ascend 950 accelerator, supporting a supernode architecture capable of linking up to 128 chips.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.
Unlocking the full article...
Contact us
Questions, corrections, or ideas? Send us a message here. We aim to reply by email within 24 hours.
Thank you. Your message has been sent. We aim to reply by email within 24 hours.Please enter your email address and a message.We could not send your message. Please try again in a little while.