Meta is releasing a new tool that it hopes will encourage developers to use the Llama family of models for their next projects. At its first LlamaCon event in Menlo Park on Tuesday, the company announced the Llama API. The tool, available as a limited free preview starting today, gives developers the ability to experiment with Meta’s artificial intelligence models, including the recently released Llama 4 Scout and Maverick systems. It also makes it easy to create new API keys that developers can use for authentication.
“We want to make it even easier for you to get started quickly with Llama, and give you full control over your models and scales without being tied to an API,” the company said in a blog post published during the event. To this end, the initial version of the Llama API includes tools that developers can use to fine-tune and evaluate their applications.
In addition, Meta notes that it will not use user prompts and model responses to train its own models. “When you’re ready, the models you build with the Llama API are yours to take with you wherever you want to put them, and we don’t keep them on our servers,” the company says. Meta plans to roll out the tool to more users in the coming weeks and months.
Despite the fact that Meta’s Llama models have been downloaded more than a billion times, the company is not generally considered a leader in the AI space like OpenAI and Anthropic. This doesn’t help fight the perception that the company was caught playing a game with LMArena to make its Llama 4 models look better than they actually were.
