[{"data":1,"prerenderedAt":147},["ShallowReactive",2],{"content-query-yDmhQGdrLJ":3},{"_path":4,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":8,"description":9,"date":10,"cover":11,"type":12,"body":13,"_type":141,"_id":142,"_source":143,"_file":144,"_stem":145,"_extension":146},"/news/en/3614","en",false,"","MindSpore Fully Supporting DeepSeek V3 Training and Inference","MindSpore open-source community has released a fine-tuning example for DeepSeek V3.","2025-02-20","https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/02/20/82a2061ebda94eb39fbc037594ba1161.png","news",{"type":14,"children":15,"toc":138},"root",[16,24,30,39,44,49,60,65,70,79,87,95,100,105,110,115,124,129],{"type":17,"tag":18,"props":19,"children":21},"element","h1",{"id":20},"mindspore-fully-supporting-deepseek-v3-training-and-inference",[22],{"type":23,"value":8},"text",{"type":17,"tag":25,"props":26,"children":27},"p",{},[28],{"type":23,"value":29},"Following the support of training and inference for DeepSeek V3 671B on Ascend clusters by MindSpore, the MindSpore version of DeepSeek V3 has also released its fine-tuning capabilities. As a result, MindSpore now fully supports end-to-end training and inference for DeepSeek V3.",{"type":17,"tag":25,"props":31,"children":32},{},[33],{"type":17,"tag":34,"props":35,"children":36},"strong",{},[37],{"type":23,"value":38},"Open Source Links",{"type":17,"tag":25,"props":40,"children":41},{},[42],{"type":23,"value":43},"MindSpore open-source community",{"type":17,"tag":25,"props":45,"children":46},{},[47],{"type":23,"value":48},"DeepSeek V3 training and fine-tuning code:",{"type":17,"tag":25,"props":50,"children":51},{},[52],{"type":17,"tag":53,"props":54,"children":58},"a",{"href":55,"rel":56},"https://gitee.com/mindspore/mindformers/tree/dev/research/deepseek3",[57],"nofollow",[59],{"type":23,"value":55},{"type":17,"tag":25,"props":61,"children":62},{},[63],{"type":23,"value":64},"Modelers community",{"type":17,"tag":25,"props":66,"children":67},{},[68],{"type":23,"value":69},"DeepSeek V3 inference code:",{"type":17,"tag":25,"props":71,"children":72},{},[73],{"type":17,"tag":53,"props":74,"children":77},{"href":75,"rel":76},"https://modelers.cn/models/MindSpore-Lab/DeepSeek-V3",[57],[78],{"type":23,"value":75},{"type":17,"tag":25,"props":80,"children":81},{},[82],{"type":17,"tag":34,"props":83,"children":84},{},[85],{"type":23,"value":86},"The links contain comprehensive step-by-step tutorials, enabling developers to get started effortlessly.",{"type":17,"tag":25,"props":88,"children":89},{},[90],{"type":17,"tag":34,"props":91,"children":92},{},[93],{"type":23,"value":94},"DeepSeek V3 MindSpore Fine-Tuning Capability Released",{"type":17,"tag":25,"props":96,"children":97},{},[98],{"type":23,"value":99},"MindSpore Transformers supports full-parameter fine-tuning of DeepSeek V3. By following these steps, you can quickly initiate the fine-tuning process on a single Atlas 800T A2 (64 GB):",{"type":17,"tag":25,"props":101,"children":102},{},[103],{"type":23,"value":104},"Environment preparation > Dataset preparation > Model weight preparation > Configuration modification > Task execution",{"type":17,"tag":25,"props":106,"children":107},{},[108],{"type":23,"value":109},"If you have any questions or suggestions while using the model, please provide feedback through one of the following communities:",{"type":17,"tag":25,"props":111,"children":112},{},[113],{"type":23,"value":114},"Discuss the use of DeepSeek V3 in the MindSpore open-source community:",{"type":17,"tag":25,"props":116,"children":117},{},[118],{"type":17,"tag":53,"props":119,"children":122},{"href":120,"rel":121},"https://gitee.com/mindspore/mindformers/issues/IBL0X5?from=project-issue",[57],[123],{"type":23,"value":120},{"type":17,"tag":25,"props":125,"children":126},{},[127],{"type":23,"value":128},"Discuss the use of DeepSeek V3 in the MindSpore forum of the Ascend community:",{"type":17,"tag":25,"props":130,"children":131},{},[132],{"type":17,"tag":53,"props":133,"children":136},{"href":134,"rel":135},"https://www.hiascend.com/forum/thread-02112174450796469017-1-1.html",[57],[137],{"type":23,"value":134},{"title":7,"searchDepth":139,"depth":139,"links":140},4,[],"markdown","content:news:en:3614.md","content","news/en/3614.md","news/en/3614","md",1776506047325]