Chief AI Sharing Circle

Voyage AI's Voyager 3 is a new state-of-the-art model that allows you to embed text and images into the same space. In this post, I will explain how to extract these multimodal embeddings from magazines, store them in a vector database (Weaviate), and use the same embedding vectors...

Synthesis SHMT (Self-supervised Hierarchical Makeup Transfer) is a self-supervised hierarchical make-up transfer project based on a latent diffusion model, aiming to achieve high-quality transfer of make-up effects through unsupervised learning methods. The project adopts the "decoupling and reconstruction" paradigm, which abandons the practice of disallowing ...

Previously, SiliconCloud went online with the online LoRA fine-tuning feature for language models. By simply uploading corpus data and creating a fine-tuning task, you can get an exclusive fine-tuned language model. Recently, SiliconCloud's LLM online LoRA fine-tuning has been extended to include Qwen2.5-32B, Qwen2.5-1...

CAG (Cache Augmented Generation) that is 40 times faster than RAG (Retrieval Augmented Generation).CAG revolutionizes knowledge acquisition: instead of retrieving external data in real time, all knowledge is pre-loaded into the model context. It's like condensing a huge library into an on-the-go toolkit that can be used when needed...

General Introduction VITA is a leading open source interactive multimodal large language modeling project, pioneering the ability to achieve true full multimodal interaction. The project launched VITA-1.0 in August 2024, pioneering the first open source interactive fully modal large language model.In December 2024, the project launched...

General Description Trend Finder is a powerful tool designed to help users track trending topics and trends on social media in real time. By collecting and analyzing posts from key influencers, Trend Finder is able to send timely Slack notifications when new trends or product releases are detected. This tool is extremely...

Currently my best AI programming partners are Lovable and Cursor. bolt.new and windsurf are also very good, I chose the first two because the ceiling is high enough. Lovable can be found at https://lovable.dev/ Lovable may not be as famous as bolt.new, but I recommend you to try it...

Yesterday, Sam Altman, CEO and co-founder of OpenAI, posted his latest in-depth article - Reflections - on his personal blog. The main review of OpenAI's founding in the past 9 years: from the initial lack of optimism, to the release of ChatGPT in 2022, which set off a global AI revolution users a...

Luo Yonghao is entering the AI industry again this time. As previously reported, his new company, Thin Red Line, will release its first new product since its inception around the Chinese New Year of the Snake. As early as last April, Luo Yonghao first teased in a live broadcast that he would release a mysterious product, and described it as "disruptive, destructive innovation...

General Introduction Matter.ai is an innovative company dedicated to providing advanced artificial intelligence solutions. Its latest product, J1 Assistant (J1 Assistant), is now available in version 0.8.3-beta1 with support for Samsung Galaxy S24 series, S23 series, S22 series as well as Pixel 9 series, Pixel 8 series...

Making predictions, especially in a fast-moving field like data and AI, is notoriously difficult. Nonetheless, we, Rajesh Parikh and Sanjeev Mohan, released our 2024 Trend Forecast last year. As 2024 draws to a close, we are pleased to confirm that our predictions are very...

TransRouter is a real-time voice translation tool based on Google's Gemini model, designed for real-time voice translation between English and Chinese. It can be seamlessly integrated into video conferencing software such as Zoom to provide real-time translation support for cross-language communication.TransRout...

Comprehensive Introduction LatentSync is an innovative audio conditional potential diffusion modeling framework open-sourced by ByteDance, specifically designed to enable high-quality video lip-synchronization. Unlike traditional approaches, LatentSync uses an end-to-end approach that eliminates the need for intermediate action representations to directly generate natural,...

General Introduction Open Source NotebookLM is an innovative AI project that combines Deepseek-V3's language understanding capabilities with PlayHT's speech synthesis technology, aiming to create an intelligent note-taking conversation system. Developed by the Build Fast with AI team, the project transforms text content into...

Comprehensive Introduction Open Deep Research is an open source AI-driven research report generation tool that serves as an open source alternative to Google Gemini's deep research capabilities. Developed in TypeScript and built on the Next.js 15 framework, the project integrates the Azure Bing Search API and Google Gemini ...

Comprehensive Introduction Vision-is-all-you-need is an innovative visual RAG (Retrieval Augmented Generation) system demo project that breaks new ground in applying Visual Language Modeling (VLM) to the document processing domain. Unlike traditional text chunking methods, the system uses visual language modeling directly to process the pages of a PDF file...

Comprehensive Introduction MiniPerplx (renamed Scira) is a minimalist designed AI-powered search engine that integrates a variety of useful features to provide users with a full range of information retrieval services. The project uses a modern technology stack, including Next.js, Tailwind CSS and Vercel AI SDK, and...

  Do you often need to transcribe meeting recordings or interviews into text? Since writing verbatim scripts is time-consuming and laborious, it's a good idea to utilize AI tools to convert audio recordings into text. In this article, we will introduce Whisper, an automatic speech recognition (ASR) system launched by the OpenAI team. According to OpenA...