💻 Running Gemma 4B Inside the ChatGPT Sandbox
User Max Jendrall demonstrated running the Gemma 4B model within the ChatGPT sandbox. To bypass restrictions when downloading from Hugging Face, ChatGPT Sites was used as a proxy. The sandbox provides ~15 GB of RAM and 9 CPU cores, allowing for the execution of autonomous tasks.
🌍 This opens up the potential for using LLM cloud sandboxes as full runtime environments for autonomous AI agents capable of deploying specialized software.
👤 ChatGPT resources can now be used to run small models or perform complex programming tasks, effectively turning the chatbot into a workstation.
Source 1: https://x.com/maxjendrall/status/2079957211216748669?s=46&t=-zZsx0fRo0EpNTv1GAtTNQ