Starting today, open source is leading the way. Introducing Llama 3.1: Our most capable models yet. Today we’re releasing a collection of new models including our long awaited 405B. Llama 3.1 delivers stronger reasoning, a larger 128K context window & improved support for 8 languages including English — among other improvements. Details in the full announcement ➡️ https://go.fb.me/hvuqhb Download the models ➡️ https://go.fb.me/11ffl7 We evaluated performance across 150+ benchmark datasets across a range of languages — in addition to extensive human evaluations in real-world scenarios. Trained on >16K NVIDIA H100 GPUs, Llama 3.1 405B is the industry leading open source foundation model and delivers state-of-the-art capabilities that rival the best closed source models in general knowledge, steerability, math, tool use and multilingual translation. We’ve also updated our license to allow developers to use the outputs from Llama models — including the 405B — to improve other models for the first time. We’re excited about how synthetic data generation and model distillation workflows with Llama will help to advance the state of AI. As Mark Zuckerberg shared this morning, we have a strong belief that open source will ensure that more people around the world have access to the benefits and opportunities of AI and that’s why we continue to take steps on the path for open source AI to become the industry standard. With these releases we’re setting the stage for unprecedented new opportunities and we can’t wait to see the innovation our newest Llama models will unlock across all levels of the AI community.
Meta for Developers’ Post
More Relevant Posts
-
📣 Meta Connect’s 2024 line-up is here! Register now and get inspired by sessions like "Llama 101", "Get Started with Mixed Reality" and more: https://bit.ly/4elsldD #MetaConnect
-
Have you seen how fast SambaNova's inference is on Llama? They are taking 405B and 70B to a new level! 👏 Learn more: https://bit.ly/3Xrju3g
-
-
Meet Somo-GPT, a revolutionary AI teaching assistant from Upeo Labs. Powered by Meta's Llama 3 and 3.1, Somo-GPT supports high school students, teachers, and parents in Kenya with a conversational chat interface on various subjects. Currently in closed beta in 500 high schools! Learn more about Upeo Labs story: https://bit.ly/3zdy9XU
-
-
Our Generative AI APIs offers 3 creative features (text generation, background generation, and image expansion) designed to enhance and scale your business’s advertising performance. Watch this video to learn more about how these features can drive increased creative diversification and efficiency in your campaigns: https://bit.ly/47fVWmF
-
-
📣 Get a front row seat to explore the latest in AI, MR, and technologies of the future. Register now: https://bit.ly/4dHtEUp
-
Pioneering a new era of AI, Llama is at the forefront of openness, customizability, and affordability. We're dedicated to developing in the open, ensuring AI benefits are accessible to all. Learn more about how you can join the growing AI community who believe open source is the key to a more inclusive future: https://bit.ly/3AG4XZL Download Llama today: https://bit.ly/4g2asCl
-
-
We're making significant changes to the way developers certify their access to our developer platform. This follows our commitment to protecting privacy and improving the developer experience. Our goal: to provide an assessment process that's simpler, easier to understand and more efficient. Learn more on our blog: https://bit.ly/4cYmtWV
-
-
We're launching the second iteration of our Meta Segment Anything Model, SAM 2, improving on the foundation of the original SAM. SAM 2 represents the next step in real-time promptable object segmentation for images and videos. We're excited by the potential for many real-world applications in arts, sciences and daily life. Learn more on our blog: https://bit.ly/4dEEXMH
-
DCPerf, our collection of benchmarks that represent the diverse categories of workloads in data center cloud deployments, is available now on GitHub. We've open-sourced with the goal of creating a reference benchmark to design, develop, debug, optimize and improve state-of-the-art in compute platform designs for hyperscale. Learn more on our blog: https://bit.ly/3X2Bthu
-
Attended Muslim College of Education
2wHello 👋.