DLite V2: Lightweight, Open, Easily Customizable LLMs

AI Squared is committed to democratizing AI so that it can be used by all. There are two key forces opposing the democratization of AI though — a tendency for high-performing models to have a huge number of parameters, making them incredibly expensive to train, tune, and deploy at scale — and nonpermissive licensing preventing many open source models from being used for commercial purposes. <img alt="" src="https://miro.medium.com/v2/resize:fit:700/0*_DXI1ZU5v2FDR88E" style="height:281px; width:700px" /> Getting high performance from smaller models would greatly reduce the start-up and operational costs of building with large language models. To address the size / cost aspect of this current situation, in April 2023 , we released our <a href="https://huggingface.co/aisquared/dlite-v1-124m" rel="noopener ugc nofollow" target="_blank">DLite V1 family of models</a>, which are lightweight LLMs ranging from 124 million parameters to 1.5 billion parameters that exhibit ChatGPT-like interactivity. The small size of these models means that they can be run on almost any device, including laptop CPUs, instead of being limited to deployment on specialized, expensive cloud resources. At this point, however, we were using the Alpaca dataset to tune the model, which prevented any of the DLite v1 family from being used for commercial purposes. <a href="https://medium.com/ai-squared/dlite-v2-lightweight-open-llms-that-can-run-anywhere-3ab59b3a7feb">Website</a>