Meta has released the initial models from its latest open-source artificial intelligence software, Llama 4.
The social media giant, which owns Facebook, has not yet released the biggest and most powerful Llama 4 model. This upcoming model, expected to outperform other AI models in its class, serves as “a teacher for our new models.”
Users can access the two newly released Llama 4 models via Meta AI in WhatsApp, Messenger, Instagram Direct, or the Meta AI website.
These models are known as Llama 4 Scout and Llama 4 Maverick.
Meta’s Chief Product Officer, Chris Cox, mentioned in March that the Llama 4 Behemoth model is still being trained. This advanced model will be capable of new levels of reasoning and action, enabling agents to surf the web and handle various tasks beneficial to both consumers and businesses.
Today is the start of a new era of natively multimodal AI innovation.
Today, we’re introducing the first Llama 4 models: Llama 4 Scout and Llama 4 Maverick — our most advanced models yet and the best in their class for multimodality.
Llama 4 Scout
• 17B-active-parameter model… pic.twitter.com/Z8P3h0MA1P— AI at Meta (@AIatMeta) April 5, 2025
Our goal is to build the world’s leading AI, open source it, and make it universally accessible so that everyone benefits,” said Meta CEO Mark Zuckerberg. And I’ve said for a while that I think open-source AI will become the leading models, and with Llama 4, this is starting to happen. Meta AI is getting a big upgrade today.”
Meta will host its first LlamaCon AI conference on April 29.
The company is also expected to announce a standalone app for its Meta AI chatbot in the second quarter.
Okay Llama 4 is def a littled cooked lol, what is this yap city pic.twitter.com/y3GvhbVz65
— Nathan Lambert (@natolambert) April 6, 2025
Meta’s new AI models launch
However, a Meta executive denied a rumor that the company tuned its new AI models to perform better on specific benchmarks while hiding the models’ weaknesses.
Ahmad Al-Dahle, VP of Generative AI at Meta, stated that it’s “simply not true” that Meta trained its AI on “test sets.
In AI benchmarks, test sets are collections of data used to evaluate a model’s performance after it’s been trained. Training on a test set could misleadingly inflate a model’s benchmark scores, making the model appear more capable than it is. Over the weekend, rumors began circulating on social platforms X and Reddit that Meta artificially boosted its new models’ benchmark results.
The speculation seems to have originated from a post on a Chinese social media site from a user claiming to have resigned from Meta in protest over the company’s benchmarking practices. Reports suggested that Maverick and Scout, Meta’s AI models, were central to the controversy. Al-Dahle acknowledged that some users are experiencing “mixed quality” from Maverick and Scout across different cloud providers hosting the models.
“Since we dropped the models as soon as they were ready, we expect it’ll take several days for all the public implementations to get dialed in,” Al-Dahle said. We’ll keep working through our bug fixes and onboarding partners.
Users have noted differences in the performance of Maverick hosted on various platforms, further complicating the narrative around its quality and benchmarks. Meta has been actively addressing these concerns and aims to resolve discrepancies in the performance experienced by some users.
Image Credits: Photo by Julio Lopez on Unsplash
Johannah Lopez is a versatile professional who seamlessly navigates two worlds. By day, she excels as a SaaS freelance writer, crafting informative and persuasive content for tech companies. By night, she showcases her vibrant personality and customer service skills as a part-time bartender. Johannah's ability to blend her writing expertise with her social finesse makes her a well-rounded and engaging storyteller in any setting.
