Google Releases Gemma 3 Family of Open-Source AI Models, Can Run on a Single GPU

Google launched the Gemma 3 household of synthetic intelligence (AI) fashions on Wednesday. Successor to the Gemma 2 sequence, which was introduced in August 2024, the brand new open-source fashions arrive with textual content and visible reasoning capabilities. The Mountain View-based tech large mentioned these fashions provide assist for greater than 35 languages and may be fine-tuned to supply extra assist for 140 languages. Notably, the tech large claimed that these fashions are optimised to run on a single GPU or Google’s Tensor processing unit (TPU).

Google Releases Gemma 3 Collection AI Fashions

In a blog post, the tech large detailed the brand new small language fashions (SLMs), highlighting that these had been constructed utilizing the identical expertise as the corporate’s flagship Gemini 2.0 fashions. The Gemma sequence is thought for being open-source and providing on-device efficiency. Google revealed that up to now, the Gemma fashions have been downloaded greater than 100 million instances and have been used to create greater than 60,000 variants.

The tech large claimed that Gemma outperforms Meta’s Llama-405B, DeepSeek-V3, OpenAI’s o3-mini AI fashions on the LMArena’s leaderboard. These fashions can be found in 4 sizes — 1B, 4B, 12B, and 27B parameters. They are often run on a single GPU or TPU, the corporate claimed.

Gemma 3 sequence affords superior textual content and visible reasoning capabilities, and may analyse pictures, textual content, and quick movies. The AI fashions provide a context window of 1,28,000 tokens. The fashions additionally provide operate calling assist, which is able to enable builders to construct agentic capabilities within the apps and software program they create.

Google mentioned that the AI fashions had been developed utilizing cautious danger evaluation. The corporate claimed that it enforced inner security insurance policies through fine-tuning and benchmark evaluations. The fashions had been additionally examined utilizing extra succesful fashions. The mannequin displayed a low danger degree, the corporate claimed.

Along with Gemma 3 sequence, the corporate additionally launched ShieldGemma 2, a 4B parameter picture security checker, that ensures that the AI fashions don’t generate content material with harmful, sexually specific, or violent content material. Builders have additionally been given the choice to additional customise ShieldGemma to boost the security parameters. Gemma 3 household of AI fashions may be downloaded through Google’s Hugging Face listing or on Kaggle.