Liquid AI introduces the first series of Liquid Foundation Models (LFMs), cutting-edge generative AI models developed from first principles. These models achieve superior performance in terms of quality, efficiency, and memory usage across different scales compared to existing models like GPT. Try LFMs on Liquid Playground, Lambda, Perplexity Labs, and soon on Cerebras Inference. LFMs are optimized for various hardware including NVIDIA, AMD, Qualcomm, Cerebras, and Apple. The Language LFMs series comprises 1.3B, 3.1B, and 40.3B models, offering efficient solutions for edge deployment and complex tasks. LFMs excel in memory efficiency, context exploitation, and model architecture to deliver state-of-the-art performance, promising innovation and scalability in AI solutions.
https://www.liquid.ai/liquid-foundation-models