GoJun AI Daily

English

简体中文

English

简体中文

Appearance

Spend 1 minute each day to get curated cutting-edge AI information.

The content covers, but is not limited to, cutting-edge AI news, AI tools, AI artwork, open-source projects, and learning tutorials.

Follow AI Daily to stay up to date with AI trends. I hope this will be helpful to you. Important information will be posted separately with detailed introductions.

Below is the latest AI news from September 4.

Cutting-edge News ​

1. Luma AI adds camera control features for video generation.

Users can input "Camera" to trigger and select different types of camera movement, such as moving left/right, zooming in/out, panning, circling, and vertical movement.

Try it out online: https://lumalabs.ai/dream-machine

The demo looks great. Now there’s another exciting way to generate videos.

AI News - September 4 Video Optimizer

2. Baidu’s Wenxin Yiyan renamed to "Wen Xiaoyan".

They also announced that the Wenxin 4.0 large model will be free to use in September.

I can’t even remember when I last used it. Is anyone still using it?

GWl-WynW0AAurmY

AI Artwork ​

1. LivePortrait, the open-source portrait animation framework from Kuaishou, now has a ComfyUI plugin!

It can animate facial expressions in a static portrait, supporting various styles of images while maintaining character consistency. The facial expression control is excellent.

More details on LivePortrait: https://liveportrait.github.io/

I found today that there’s a ComfyUI plugin for it on GitHub: ComfyUI-AdvancedLivePortrait.

It’s faster and offers real-time preview. You can directly edit facial expressions in images and insert them into videos. I believe this will make for some fun video creations.

GitHub: https://github.com/PowerHouseMan/ComfyUI-AdvancedLivePortrait

Now available for installation via ComfyUI-Manager. Give it a try if you’re interested.

ComfyUI AdvancedLivePortrait

Open-source Projects ​

1. An open-source AI code editor designed for 10x engineers: Melty.

It integrates with the entire development workflow, understanding everything from your terminal to GitHub operations, helping you write and refactor code more efficiently. It can also perform large-scale changes across multiple files.

GitHub: https://github.com/meltylabs/melty

Its main goals are:

  • Helping developers better understand and maintain code.
  • Observing every change you make like a pair programmer.
  • Learning and adapting to your codebase.
  • Integrating with compilers, terminals, debuggers, as well as tools like Linear and GitHub.

Quick tip: 10x engineers are those who are 10 times more efficient than others.

Charlie Holtz coding differently

2. A curated list of resources related to retrieval-augmented generation (RAG) in large language models: Awesome-RAG.

It covers nearly four years of representative research papers on RAG, as well as related lectures, seminars, tutorials, tools, and other resource collections.

GitHub: https://github.com/coree/awesome-rag

image-20240904134418102

3. An open-source, free, fully automated AI video editing tool: MoneyPrinterPlus.

With AI, it allows you to generate and batch-edit various short videos with one click, and automatically publish them to multiple video platforms, fully automating the process to help users monetize content.

GitHub: https://github.com/ddean2009/MoneyPrinterPlus

Main features:

  • Supports local speech models like ChatTTS, FasterWhisper, GPTSoVITS, etc.
  • Supports batch video auto-publishing to various platforms.
  • Supports batch video editing, producing non-repetitive short videos.
  • Allows choosing local materials and supports various resolutions.
  • Supports local models like Ollama or API service models.
  • Supports over 100 different voice types and speed adjustments.
  • Supports 30+ transition effects and subtitle effects.

image-20240628153020140