Ahh I see. What size model in your opinion do they get smarter? 70B? Or do you need much larger? Thanks for all the input and replies. I appreciate it.

What do you mean by “not smart enough for openclaw”? I didn’t think openclaw used that many resources to run. I’m sure your usage is heavier than most and you utilize a lot of compute.

Will this allow more power to the home user trying to run an open claw type application? Will this help run local open source LLM’s or are these two things unrelated?