🚀 Exciting AI Updates: Unleash the Power of LLMs Offline for Document Queries | Discover the Latest Advancement in LLMs with Consumer GPUs | Introducing Stability AI's Open-Source Text-to-Animation Tool.....

Stanford Researchers Propose FrugalGPT: a simple yet flexible instantiation of LLM cascade which learns which combinations of LLMs to use for different queries in order to reduce cost and improve accuracy. According to the research paper, FrugalGPT can match the performance of the best individual LLM (e.g. GPT-4) with up to 98% cost reduction or improve the accuracy over GPT-4 by 4% with the same cost.

Do you know what's cooler than running LLMs on consumer GPUs? Finetuning large 65B+ LLMs on consumer GPUs! Meet LLMTune: 4-Bit Finetuning of LLMs on a Consumer GPU. LLMTune allows finetuning LLMs (e.g., the largest 65B LLAMA models) on as little as one consumer-grade GPU. Its features include: Modular support for multiple LLMs (currently LLAMA, OPT), Support for a wide range of consumer-grade NVidia GPUs; 65B LLAMAs finetune on one A6000 and Tiny and easy-to-use codebase.

Meet Dromedary: A bootstrap open-sourced LLM model to align itself with minimal human supervision. It’s a fresh breeze during the time when everyone just scrapes and imitates OpenAI’s outputs. Never underestimate the power of self-hosting. With fewer than 300 lines of human annotations (including < 200 seed prompts, 16 generic principles, and 5 exemplars for in-context learning), Dromedary significantly surpasses the performance of several state-of-the-art AI systems, including Text-Davinci-003 and Alpaca, on benchmark datasets with various settings.

Stability AI releases an open-source text-to-animation tool. The Stable Animation SDK empowers users to create captivating animations by combining text, images, or videos with text. This versatile tool enables the generation of videos through three distinct input options: text alone, text with an initial image, or text with an input video. Unlike the accessible web-based platforms like DALL-E or Bing Image Creator, accessing this model demands a higher level of technical expertise to install and operate the user interface, involving coding skills.

Meet Blendify: A Python Framework Developed with a Focus on 3D Computer Vision Visualization. Blender is a popular open-source 3D creation software that is used for modeling, animation, and rendering. Developed by Ph.D. students Vladimir Guzov and Ilya Petrov, Blendify aims to simplify the process of using Blender for computer vision visualizations by providing a user-friendly interface without the need to interact directly with the Blender API. The user doesn’t have to delve into the complexities of the Blender API to achieve the desired visualizations.

Scale AI just launched 2 major platforms to bolster government and enterprise. The first platform is called Scale Donovan, the AI copilot for defense. While the second platform is Scale EGP, a full-stack generative AI for the global enterprises. Donovan is an AI-powered platform to help operators understand, plan, and act in minutes instead of weeks. Scale EGP is a full-stack solution for any company looking to deploy generative AI applications in production.

Meet privateGPT: Ask questions to your documents without an internet connection using the power of LLMs. 100% private; no data leaves your execution environment at any point. You can ingest documents and ask questions without an internet connection! Yes, you heard it right.