DeepSeek and Huawei have jointly released a set of open-source programming tools for Huawei's Ascend AI chips, aiming to reduce the industry's reliance on Nvidia's software and hardware ecosystem. The announcement, reported by Reuters on September 30, includes libraries for AI computation and inter-chip communication, alongside native support for the high-level programming language TileLang on Ascend 950 processors.
The two companies said Huawei provided full support during development. The tools are intended to simplify programming while allowing developers to fully exploit the hardware's performance. The work addresses two critical requirements for running large AI workloads across multiple accelerators: performing calculations efficiently on each chip and moving data quickly enough between them to keep the processors occupied.
Among the released libraries is DeepGEMM-Ascend, which handles matrix multiplication and other calculations used in DeepSeek's models. It supports BF16, FP8, and FP4 operations and uses the same programming interfaces as DeepSeek's existing DeepGEMM library, allowing developers to retain familiar APIs when moving to Ascend. DeepEP-Ascend handles communication for training and inference, including routing data to the different experts in mixture-of-experts models and combining their outputs. Both libraries were developed and tested on Ascend 950 hardware.
TileLang provides a higher-level programming layer for writing optimized kernels. DeepSeek described it as offering "a simpler programming model" than Nvidia's CUDA, with the aim of improving development efficiency and simplifying code. The language already supported Nvidia and other hardware, with earlier adapters available for Huawei's Ascend processors. The September 30 update adds native support for Ascend 950, including code generation, automatic scheduling, and synchronization.
The tools build on Huawei's existing CANN software platform, which provides the underlying infrastructure for running AI workloads on Ascend. Nvidia's CUDA platform has long supplied developers with a mature programming environment and libraries optimized for its GPUs, making the software ecosystem a major part of the company's advantage in AI computing. While DeepSeek's release provides developers with additional tools to optimize workloads on Huawei hardware, TileLang's support for multiple platforms ensures the language remains useful for Nvidia GPUs.
The announcement comes two weeks after Huawei unveiled its next generation of AI processors and supernode systems. Huawei said it expected its AI systems to be widely used for model training in 2027. The companies had already collaborated on DeepSeek's V4 model, released in preview form in April, with support for Huawei's Ascend chips. Huawei said its Ascend 950 supernodes fully supported the V4 models and that its chips had been used for part of the lighter V4-Flash variant.