China Telecom's Xing4 Report: A 285B-Parameter Model Trained on 8,192 Huawei Ascend Chips
Posted on 18th Nov 2026 11:03:38 in Artificial Intelligence, Machine Learning
Tagged as: China Telecom, Huawei Ascend, Xing4, AI chips, MindSpore, open weights
China Telecom's Xingchen lab has published the training stack behind Xing4, including a 285B-parameter model pretrained on 8,192 Huawei Ascend 910C chips with MindSpore - no Nvidia GPUs, no CUDA - alongside an Apache 2.0 29B agentic model developers can run today.
Mistral Large 4 Le Chonk: Europe's Biggest Open-Weight Model Yet Bets on Cyber Defense
Posted on 7th Oct 2026 12:05:23 in Artificial Intelligence, Machine Learning
Tagged as: Mistral, Le Chonk, Mistral Large 4, open weights, cybersecurity, AI models, Europe
Mistral has opened a public preview of Mistral Large 4, a 1-trillion-parameter open-weight model nicknamed Le Chonk that it says is the strongest built outside China - trained entirely in Europe and tuned for cyber defense, with weights due October 27.