Huawei Open-Sources openPangu-2.0 Training Code for Ascend
Huawei Open-Sources openPangu-2.0 Pretrain, SFT and RL Training Code on Ascend
Training code was the missing piece after the weights shipped, so this makes openPangu one of the few large MoE models reproducible end to end on non-NVIDIA silicon. Teams evaluating Ascend for large-scale training, or comparing Chinese open-source stacks against Qwen and DeepSeek, can now audit the full recipe rather than only run inference.
Huawei released the pretrain, SFT and post-training RL code of openPangu-2.
0 as open source on 28 Sep 2026.The code is built for Huawei's Ascend-native MoE technology stack.
It complements the previously released openPangu-2.0 Pro (505B) and Flash (92B) model weights.
Developers can now reproduce the full training pipeline on Ascend hardware.
Huawei Open-Sources openPangu-2.0 Pretrain, SFT and RL Training Code on Ascend
On September 28, 2026 Huawei open-sourced openPangu-2.0 pretrain, SFT and post-training RL code for its Ascend-native MoE stack, complementing earlier Pro (505B) and Flash (92B) weight releases.
This report is categorized under AITOP's "Models" desk, focusing on China's AI industry and domestic foundation model ecosystem. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows and US-China foundation model benchmarks and commercialization comparison. Key entities and technical topics under continuous tracking include Huawei, openPangu, Ascend, open-source models.
Primary source verification: published directly by pandaily with first-party authority. Cross-source corroboration: 7 independent sources have covered this development (including IT之家, Gorden Sun, AIGCLINK), confirming industry-wide signal strength.
