Mythos Preview is currently available to our launch partners in Project Glasswing. Learn more about the model and the project here: https://www.anthropic.com/glasswing this will change our entire workflow, also our bank accounts 🤣

We are just 21 days in since starting this journey in the community. But the vision has not changed. 202 members joined all with a similar vision. How can we use AI to grow as people, grow our businesses, and to provide value back into the economy. The focus of the community has always been us growing together, sharing ideas, and helping us learn new skills in this AI age. I will keep providing as much value as I can, adding more to the course material and resources pages. Bringing on new coaches to teach and show off validated openclaw use cases. But I also appreciate your feedback. If you have anything you would like to see in the future, please post under here. Do you want more workshops? Different types of workshops? What would you like to see in the course material? Just wanted to show my appreciation and say I'm so thankful for everyone here. You're in the right spot if you want to learn how to use AI. It's not too late. It's just the right time! UPDATE: many requests for a beginner's basic OpenCLaw class, like getting OpenCLaw installed, beginner basics, OpenCLaw principles. @Bret Jutras is going to be doing another class going forward in addition to what we have already. It's basically going to be the very basics of how to get OpenCLaw installed, what models to choose, and how to get it to work without breaking. Appreciate you all so much and have a blessed Easter weekend.

Claude closed access to 3rd Party harnesses. Well Gemma 4 might be your solution, it looks like mine. Thoughts?

I've been testing all day with new models to be able to save some money. I think we're all trying to do that. Tried Kimi K 2.5 Tried GPT 5.4. Tried Minimax 2.7 then i tried GLM 5.1(PRO) And my goal is to not use it for coding but just as the execution and coordination agent that I talk back and forth with. My main agent. All the other models seemed very generic, with no personality, and that bothered me. But I finally found something today in GLM 5.1 and it's hard to even tell the difference between Sonnet and GLM 5.1. So if you're on a budget and you're looking for a model that doesn't hallucinate and go crazy and forget who you are and break things, I say give GLM 5.1 a try as your main agent. I paid for the $30/month plan, which has a 210,000 context token window. It performs very well for my use case. again im not using it for coding. just coordination/execution. hope this helps

Its over. You will have to use api or switch to gpt 5.4 I have been using gpt 5.4 all week and its been great! my plan is pretty simple going forward: - keep gpt-5.4 running most of the cron jobs and backend work - use claude more selectively where the quality really matters - install claude code locally on my mac mini so it can review and fix my openclaw setup directly on the same machine - keep testing the best balance between quality and cost basically:openclaw keeps running the machine claude code helps me audit and fix the machine and i stop wasting money where i dont need to should be a cleaner setup and probably a cheaper one too lol