https://lxe.co/wasmgpt/ you can now run a LLM in your browser.
This is absolutely amazing. You download a 900MB neural network to your smartphone by simply visiting a website and you can chat to it - offline. Web proves itself as the ultimate software distribution platform. Yes, the model is not very smart, but it just works, it's private and it's a stunning demo of what is to come.
We are living in the future.
#chatgpt #ai #llm #textgeneration #chatbot #Assistant #machinelearning
https://huggingface.co/spaces/togethercomputer/OpenChatKit OpenChatKit is an open source foundational large language model that you can use to build your own chat bots.
The best thing is you can try it out now! It's free, no registration needed, click agree terms and services and start using it
... and even better: help them improve their model by 👍 👎 responses
It reacts really fast and can answer very concise, which makes it feel like a real conversation.
https://github.com/Significant-Gravitas/Auto-GPT "Auto-GPT is [...] driven by GPT-4, chains together LLM "thoughts", to autonomously achieve whatever goal you set.
From the docs:
"💀 Continuous Mode ⚠️
Run the AI **without** user authorization, 100% automated. Continuous mode is NOT recommended."
AutoGPT will attempt to write and improve it's own code, debug it, and repeat, you as user can guide it or go fully auto 😱
https://open-assistant.io/chat Open-Assistant was just release.
It's like ChatGPT from OpenAI but it's actually open(source) 🎉. It has been trained on the largest human interaction dataset in existence, created by the community since around last Christmas 🙏.
You can access it for free and so far with my (very limited) experience can say it is on the level with ChatGPT
https://github.com/audapolis/audapolis transcribe audio and video to text, then edit the text to remove "eeems", "uuuh" (see the ζ character in the screenshot) or even full sentences - finally export the edited video/audio
This free and open source software works on all major OSes, downloads the ML language models automatically, after a few seconds of initialization the transcription works really fast ... feels like 🪄🧙♂️
#adobe #podcast #AI #videoediting #audioediting #speechtotext #transcription #audio #video
https://cocktailpeanut.github.io/dalai/#/ a dead simple installer for 🦙 (LLaMA Facebooks language model). Yes you can now run your own "ChatGPT" on your computer. 🤩
When you run the installer make sure you have enough free RAM or it will crash on you!
It runs even if you don't have a great graphics card but expect your CPU to be on 🔥
Ok, next try. @falko told me that I need to provide more information. I designed and built a 3D printable, working replica of the Titanic's engine from old blueprints that I found in books and on the internet. She was meant to be printed at an scale of 1:75 and run on a small compressor but I couldn't get her to run due to friction issues so I reprinted everything at a scale of 1:37 and now I can hook up my vacuum cleaner to the exhaust and she just flies.
TypeScripting the technical interview
https://www.richard-towers.com/2023/03/11/typescripting-the-technical-interview.html
✉️ Have you ever tried to send a HTML email? Your experience was not great? 😆
Have a look at this
https://www.caniemail.com/scoreboard/
The CSS support of different email clients varies so much that you basically have to fall back to 1995 CSS/HTML ... design everything in tables 😣
Oh and who is at the bottom of the table? I'm sure no one uses that insignificant client, no need to support that right? ... right? 😭
Machine Revolution
olivier.berlin
richardTowers
