Open Source LLM Comparison – Is Opus Cooked?

I'm moving off of Claude Code due to constant limit hits. I use it for some personal software development, but I mostly use it for simple personal assistant queries via the Hermes harness and signal.

As I was moving off, I wanted a quick and easy way to compare what Opus would produce with what some open solutions can do. So I spun up a little comparison markdown and generated a webpage from it. I'll update it as new open models come online and I can hit them from OpenRouter.

Right now I'm really jazzed about how cheap StepFun 3.5 Flash for these basic personal assistant tasks that feel like the really valuable part of all of this.

Is Opus cooked? For complex engineering work I'm still unsure, but for front-end design I think it still comes out ahead. I do think you can get a digital assistant that when wired up with the right harness, you could probably run for under ~$5 a month.