Google announced Wednesday that it will make some of its artificial intelligence models available to outside developers. The Gemma models are available to businesses as well as individuals and are free, the company said in a blog post.

The tech behemoth said it is also releasing a set of tools so developers who use the Gemma models can innovate, collaborate and work with them responsibly.

Gemma enters the market as several other AI companies, including Meta, introduce similar open systems, raising fears about enabling greater access. On Wednesday, the Biden administration said it was seeking public input about making AI so widely available.

Built around the same technology and research Google used to develop its larger AI model, Gemini, Gemma shares some features while making the technology easier to use, such as the ability to run on a developer’s laptop computer or desktop. While smaller than Gemini, Google designed the Gemma models to perform reasoning and other tasks more similarly to a larger AI program, the company said.

Gemini is comparable to OpenAI’s popular ChatGPT and a growing cohort of other AI systems, which both understand and create texts.

Gemma uses smaller sets of data known as model weights to recognize patterns and generate responses. Gemma’s model weights use two billion or seven billion parameters — far fewer than OpenAI’s GPT-3 model which uses about 175 billion or Meta’s Llama 2 models which use up to 70 billion parameters.

In October, President Joe Biden signed an executive order that gives U.S. Commerce Secretary Gina Raimondo until this summer to make recommendations about how to manage AI’s risks and benefits.

“There can be substantial benefits to innovation, but also substantial security risks, such as the removal of safeguards within the model,” Biden said in last year’s executive order.