During the 2025 Huawei China Partner Conference, Huawei announced that it has open-sourced the code of its ModelEngine AI full-process toolchain for global developers. This toolchain aims to provide solutions for the time-consuming data engineering, model training, and application deployment challenges faced during the implementation of AI in various industries.

The open-source version of ModelEngine establishes a comprehensive AI toolchain covering data enablement, model enablement, and application enablement. Its goal is to shorten the overall cycle from general large models to industry-specific models trained on private domain data, and further to the development of AI applications, thereby accelerating the adoption of AI technologies across sectors.

In terms of data enablement, the toolchain offers aggregation and processing functions for various types of data, including text, images, and documents. It provides key capabilities such as data cleaning, evaluation, QA pair generation, and knowledge creation, offering essential corpus and knowledge support for model training and RAG applications.

For model enablement, the open-source version of ModelEngine supports model deployment, training, fine-tuning, and inference, enhancing the efficiency of model training and reasoning through one-click operations.

In application enablement, this toolchain provides a one-stop development, debugging, and publishing environment for AI applications, facilitating rapid construction and validation of AI application outcomes.

Furthermore, the source code of the ModelEngine open-source version has been hosted on multiple platforms. Tech enthusiasts are welcome to share feedback and suggestions within the community to jointly promote the improvement and growth of ModelEngine, fostering a more vibrant AI open-source ecosystem.