Chinese AI firm DeepSeek has data-analytics-id=”inline-link” href=”https://api-docs.deepseek.com/news/news250929″ target=”_blank” data-url=”https://api-docs.deepseek.com/news/news250929″ referrerpolicy=”no-referrer-when-downgrade” data-hl-processed=”none”>released its latest large language model, DeepSeek-V3.2-Exp, with first-day optimizations for Huawei’s Ascend hardware and CANN software stack. The launch marks a shift in priorities to ensure leading-edge models run on domestic accelerators rather than data-analytics-id=”inline-link” href=”https://www.tomshardware.com/tech-industry/artificial-intelligence/deepseek-reportedly-urged-by-chinese-authorities-to-train-new-model-on-huawei-hardware-after-multiple-failures-r2-training-to-switch-back-to-nvidia-hardware-while-ascend-gpus-handle-inference” data-before-rewrite-localise=”https://www.tomshardware.com/tech-industry/artificial-intelligence/deepseek-reportedly-urged-by-chinese-authorities-to-train-new-model-on-huawei-hardware-after-multiple-failures-r2-training-to-switch-back-to-nvidia-hardware-while-ascend-gpus-handle-inference”>relying on Nvidia’s CUDA ecosystem.
DeepSeek announced the model on September 29, posting code and checkpoints to Hugging Face alongside a technical report. The company describes V3.2-Exp as an “intermediate step toward…
