Gemma
lastmod 2026-04-04created 2026-04-037 edits
(It's different from gamma! From gamma!)
Gemma is a series of open-source large language models developed by Google DeepMind. It is based on technology similar to Gemini. The first version was released in February 2024, followed by Gemma 2 in June 2024 and Gemma 3 in March 2025. Variants of Gemma have also been developed, such as PaliGemma, a vision-language model, and DolphinGemma, for understanding dolphin communication.
An Open-Weights1 lightweight large language model (sLLM) released by Google. It is a lightweight version of Gemini.
Gemma 4
See also
Favorite site
References
-
Because only the weights are released and not the training data, it does not meet the definition of open source set out by the Open Source Initiative (OSI). Whereas LLaMA and DeepSeek, which have a similar structure, market themselves as open source, Gemma honestly promotes itself as an Open weights LLM, as also stated on GitHub ↩