Scholar Puyu 20B Model Launched in Alibaba Cloud's MoDa Open Source, Scholar and MoDa Establish Ecological Cooperation

On September 20th, Shanghai Artificial Intelligence Laboratory and other institutions released the 20 billion parameter version of the InternLM 20B, which is open source and commercially available on the ModelScope of Alibaba Cloud. The Shusheng Puyu Big Model System and the Magic Building Community have established a significant ecological cooperation to jointly promote the ecological construction of China's Big Model

On September 20th, Shanghai Artificial Intelligence Laboratory and other institutions released the 20 billion parameter version of the InternLM 20B, which is open source and commercially available on the ModelScope of Alibaba Cloud. The Shusheng Puyu Big Model System and the Magic Building Community have established a significant ecological cooperation to jointly promote the ecological construction of China's Big Model.

The InternLM big language model was jointly launched by Shanghai Artificial Intelligence Laboratory and multiple institutions. In June this year, the InternLM 100 billion parameter (104B) language model was first released and has undergone multiple rounds of upgrades; In July, Shanghai Artificial Intelligence Laboratory opened up the lightweight version of InternLM-7B with 7 billion parameters from Puyu Scholar, and took the lead in the industry in opening up a full chain tool system that runs through data, pre training, fine-tuning, deployment, and evaluation. InternLM-7B has also been launched in the Magic Tag community.

The InternLM-20B released this time is a medium-sized large model with advanced performance and convenient application. The model is based on 2.3TTokens pre trained corpus and trained from scratch. Compared to InternLM-7B, its understanding, reasoning, mathematical, and programming abilities have significantly improved.

Compared to the 7B and 13B specification models that have been successively opened up in China before, the 20B level model has stronger comprehensive capabilities, particularly outstanding complex reasoning and reflection abilities, and can provide more powerful performance support for practical application scenarios; At the same time, the 20B level model can be inferred on a single card, and after low bit quantization, it can run on a single consumer grade GPU, making it more convenient in practical applications.

The Shusheng Puyu open source tool chain has also been completely upgraded, forming a more complete system, including the pre training framework InternLM Train, low-cost fine-tuning framework XTuner, deployment inference framework LMDeploy, evaluation framework OpenCompass, and scenario oriented intelligent agent framework Lagent. The Shusheng Puyu toolchain will form a powerful open-source tool and data system with the open-source data platform OpenDataLab, providing full chain research and application support for the industry.

The Mojia Community has opened a "Model Brand Library" page for Scholar Puyu, which aggregates all models and experience interfaces of the Scholar Puyu series, making it easy for developers to query, download, and use Scholar models in a one-stop manner; Magic official account launched a best practice tutorial to run through the deployment, reasoning and fine-tuning process of the model in advance for developers' reference.

Alibaba Cloud Magic Building Community is an AI model community with over 2 million developers, gathering over 1000 high-quality AI models contributed by more than 20 leading AI institutions. It provides developers with a one-stop model experience, download, inference, optimization, customization, and other services. The cumulative download volume of community models has exceeded 75 million times.

The original intention of jointly building the Shusheng Puyu Big Model System and the Magic Building Community is to promote the ecological prosperity of China's big model through open source and openness. Shanghai Artificial Intelligence Laboratory adheres to the principle of open source and comprehensive empowerment of AI community ecology, supporting academic research and industrial development. Alibaba Cloud, on the other hand, regards promoting the prosperity of China's big model ecosystem as its primary goal, leading the construction of China's largest AI model open-source community and providing comprehensive cloud services for big model enterprises and developers.

Disclaimer: The content of this article is sourced from the internet. The copyright of the text, images, and other materials belongs to the original author. The platform reprints the materials for the purpose of conveying more information. The content of the article is for reference and learning only, and should not be used for commercial purposes. If it infringes on your legitimate rights and interests, please contact us promptly and we will handle it as soon as possible! We respect copyright and are committed to protecting it. Thank you for sharing.(Email:[email protected])