Lambda
6 days ago
Lambda Unicorn News - September 30, 2024
Lambda introduces a new generation of generative AI with its Liquid Foundation Models, designed for efficiency and compatibility with major tech hardware. These models promise top-tier performance across various data types, pushing the boundaries of AI technology.
() Introduction
() Announcing Liquid Foundation Models: A New Generation of Generative AI
Lambda has unveiled its latest advancement in AI technology, the Liquid Foundation Models (LFMs), which are available in configurations of 1 billion, 3 billion, and 40 billion parameters. These models are designed to offer high performance with reduced memory usage and increased efficiency. The LFMs can be accessed through various platforms, including Liquid Playground, Lambda's Chat UI and API, and Perplexity Labs, with future availability on Cerebras Inference. They are compatible with hardware from major tech companies such as NVIDIA, AMD, Qualcomm, Cerebras, and Apple. The technology behind these models incorporates elements from dynamical systems, signal processing, and numerical linear algebra, enabling efficient processing of diverse data types like video, audio, text, and time series. This makes the LFMs highly versatile for a wide range of AI applications.
