{"type":"rich","version":"1.0","provider_name":"Transistor","provider_url":"https://transistor.fm","author_name":"AI News Today | Julian Goldie Podcast","title":"Minimax M3 + Hermes Agent is Insane (FREE!)","html":"<iframe width=\"100%\" height=\"180\" frameborder=\"no\" scrolling=\"no\" seamless src=\"https://share.transistor.fm/e/a59b3e62\"></iframe>","width":"100%","height":180,"duration":448,"description":"MiniMax M3 + Hermes Agent: Free 12-Hour AI Worker on Your Computer (Setup + Live Tests)\nThe video introduces MiniMax M3, a new agentic and coding model from China, and shows how to connect it to Hermes Agent to create a free AI worker that can run autonomously for up to 12 hours, controlling apps, searching the web, scheduling tasks, and sending reports. It walks through setup using Hermes Agent with Ollama running in the background and selecting MiniMax M3, then demonstrates live agent tasks like opening ChatGPT in a browser and creating a note in the local Notes app. It notes potential speed limits when using Ollama’s cloud model and suggests using MiniMax’s coding plan for faster performance via OAuth. The video also previews a cloud-hosted Hermes option (Max Hermes) powered by M3 and explains the “brain (M3) + body (Hermes)” pairing, pricing, benchmarks, and agentic capabilities.\n00:00 Minimax M3 Launch\n00:36 Benchmarks and Setup\n01:09 Install Hermes and Ollama\n01:21 Connect Model in Terminal\n02:12 Live Agent Task Demos\n02:52 Speed Tips and Plans\n03:21 Max Hermes Cloud Option\n04:15 Brain vs Body Explained\n05:33 Performance and Open Weights\n06:04 Wrap Up and Resources\n06:41 Community and Full Guide\n07:27 Final Thanks","thumbnail_url":"https://img.transistorcdn.com/Tp0JSUlyiAe2ran44b13cjk4ImQYS1QEMKCAa9Q7go0/rs:fill:0:0:1/w:400/h:400/q:60/mb:500000/aHR0cHM6Ly9pbWct/dXBsb2FkLXByb2R1/Y3Rpb24udHJhbnNp/c3Rvci5mbS9lNTYz/NGQxMTU2ZTM1MzU5/MDNiYjcwZjJmYTY5/ODJmOS5qcGc.webp","thumbnail_width":300,"thumbnail_height":300}