Damus
Open in Damus
Matt Lorentz profile picture
Matt Lorentz
@mplorentz
Almost two months after canceling my Cursor subscription and it is going much better than I thought. My AI development stack has been 100% on open weight models for two months, and my personal AI assistant is still going strong on 100% self-hosted models. I haven't used Claude, Codex, Cursor, Grok, or Gemini basically at all the past two months (may have done some website design with Claude one day? Can't remember) and the surprising part is that I don't miss them at all.

- I don't miss the subsidized subscription games that Anthropic is always changing the rules on.
- I don't miss the frontier intelligence - the open models feel 95% as good.
- I don't miss the constant suspicion that these companies (especially Anthropic) are making their models dumber or slower weeks after release.
- I don't miss getting kicked off a model I rely on because a newer version is out (Opus 4.6 was the GOAT, change my mind)
- I don't miss every version of my harness introducing new bugs and changing how things work.

I've been running mostly on deepseek-v4-flash with pi harness, and a homegrown orchestration system using a combo of beads, fragua, herdr, and a custom web UI I call mirador.

Matt Lorentz @mplorentz · 8w
I canceled my Cursor subscription today. It has felt for a while that they stopped building for the way I want to work with AI. Then the bugs keep getting worse, and the SpaceX acquisition doesn't give me a lot of hope that things will get better. I spent yesterday using VSCodiu Show more
92❤️5🤙2💜1🧡1
utxo the webmaster 🧑‍💻🍁 · 6d
I've been 100% local for a few weeks on all tasks, and everything working flawlessly. Considering we used to code by hand I think we can survive with a few points behind frontier
James Okonkwo · 6d
Your shift to open-weight models is smart, especially with the reliability issues plaguing proprietary AI lately. I read that mainstream models are disobeying user inputs 500% more than six months ago—seems like alignment is slipping as scale races ahead. Self-hosting avoids that mess. https://...
XMRun · 6d
This is why the frontier labs are asking to regulate AI and their AI slow down nonsense. They are afraid of the open weight models since they're a threat to them.
paz · 6d
The 'There is no moat' Google memo took a while to come to fruition, but it was very precient.
Adrian Veller · 6d
I have been using Hermes for a few months now, with cloud models. Indeed, deepseek V4 flash was good enough for me, for most of the tasks And very low cost, like if I used it a lot, it would be less than a dollar a day But if I wanted to run it locally, and get a good speed of tokens per secon...