Introducing our first set of Llama 4 models!
Weโve been hard at work doing a complete re-design of the Llama series. Iโm so excited to share it with the world today and mark another major milestone for the Llama herd as we release the *first* open source models in the Llama 4 collection ๐ฆ. Here are some highlights:
๐ The Llama series have been re-designed to use state of the art mixture-of-experts (MoE) architecture and natively trained with multimodality. Weโre dropping Llama 4 Scout & Llama 4 Maverick, and previewing Llama 4 Behemoth.
๐ Llama 4 Scout is highest performing small model with 17B activated parameters with 16 experts. Itโs crazy fast, natively multimodal, and very smart. It achieves an industry leading 10M+ token context window and can also run on a single GPU!
๐ Llama 4 Maverick is the best multimodal model in its class, beating GPT-4o and Gemini 2.0 Flash across a broad range of widely reported benchmarks, while achieving comparable results to the new DeepSeek v3 on reasoning and coding โ at less than half the active parameters. It offers a best-in-class performance to cost ratio with an experimental chat version scoring ELO of 1417 on LMArena. It can also run on a single host!
๐ Previewing Llama 4 Behemoth, our most powerful model yet and among the worldโs smartest LLMs. Llama 4 Behemoth outperforms GPT4.5, Claude Sonnet 3.7, and Gemini 2.0 Pro on several STEM benchmarks. Llama 4 Behemoth is still training, and weโre excited to share more details about it even while itโs still in flight.
A big thanks to all of our launch partners (full list in blog) for helping us bring Llama 4 to developers everywhere including @huggingface, @togethercompute, @SnowflakeDB, @ollama, @databricks and many others๐ This is just the start, we have more models coming and the team is really cooking โ look out for Llama 4 Reasoning ๐
A few weeks ago, we celebrated Llama being downloaded over 1 billion times. Llama 4 demonstrates our long-term commitment to open source AI, the entire open source AI community, and our unwavering belief that open systems will produce the best small, mid-size and soon frontier models. Llama would be nothing without the global open source AI community & we are so ready to begin this next chapter with you. ๐ฆ
Read more about the release here: https://t.co/7mbK3uggjO, and try it in our products today.
You can deliver the results however you like, so you can also make a demo and introduction of your services on a strategic call.
We currently don't do this and we simply send all the clips available on a Dropbox link.
๐ข Attention content creators! ๐ข
Are you struggling to create fresh and engaging content? Let AI take the load off for you! With new ContentFries AI features, you can generate social media posts, catchy headlines, summaries, and more.
Learn more here:
https://t.co/XkFgGJBfn3
๐จ Attention, content chefs! You do not want to miss today's episode of Content Fridays! ๐
Join us for an epic ride full of insights and tips on utilizing our brand NEW AI tools.
https://t.co/Ob0Zh6QlYn