Huawei open-sources a 505B model trained entirely on its own chips — no Nvidia anywhere
On July 31 Huawei released the weights, inference code and technical report of openPangu-2.0-Pro, a 505-billion-parameter mixture-of-experts model (about 18B active per token) with a 512K context window, trained on roughly 34 trillion tokens. The distinctive part is the hardware: the entire training run happened on Huawei's own Ascend NPUs, with no Nvidia chips involved. The release fulfils a promise Huawei made at its developer conference in June.
Why it mattersUS export controls were meant to keep China from training frontier-scale models — and here is a 505B open model built wholly on domestic silicon, with the recipe published. For the open-source world it is one more heavyweight to build on; for policymakers it is hard evidence that the chip wall is leaking.
Read in the app — free, in 9 languages
Related stories
Reflection will hand out a 501-billion-parameter model for free.2026-10-06Microsoft's new transcriber starts writing before you finish speaking.
2026-10-05An American lab is about to give away a China-class open model.
2026-10-05A video site now gives away the best-scoring open translation model.
2026-10-05Half the people on a short video call thought the AI was human.
2026-10-04