Google recently announced that the download count for its AI model suite, Gemma, has surpassed 150 million, marking significant progress for the project. Omar Sanseviero, a developer relations engineer at Google DeepMind, shared this news on social platform X and revealed that developers have created over 70,000 different versions of the Gemma model on the AI development platform Hugging Face.

QQ_1747097400799.png

Gemma was officially released in February 2024 with the aim of competing with other open model series like Meta's Llama. Unlike many existing models, Gemma is multimodal, meaning it can handle not only text but also images, expanding its range of applications. Additionally, Gemma supports more than 100 languages, meeting the needs of global users. Google has also introduced fine-tuned versions of Gemma for specific application areas, such as drug research, to enhance its applicability in professional fields.

QQ_1747097465528.png

Despite reaching 150 million downloads in just one year, this number still falls far short of its main competitor, Llama. According to statistics, by the end of April, Llama's download count had exceeded 1.2 billion. Notably, both Gemma and Llama have faced criticism from some developers due to their customized, non-standard licensing terms, which they believe pose certain risks when using these models in commercial environments.

The competition between Google’s Gemma and Meta’s Llama reflects the rapid development and intense market competition in today’s AI field. As more developers enter this domain, we can expect to see more innovative applications and solutions emerging in the future.

Key Takeaways:

🌟 Breaking 150 Million Downloads: Google’s Gemma AI model suite has reached a new milestone, attracting significant attention from developers.

🖼️ Multimodal Functionality: Gemma not only handles text processing but also image processing, supporting over 100 languages, enhancing its versatility.

⚠️ Licensing Term Controversy: Both Gemma and Llama have faced criticism due to their complex licensing terms, with concerns about risks in commercial use.