Building Auto Mode for Open Models

I can be your angle... or yuor devil.

Open-source models are getting too good to ignore. The latest Kimi, Deepseek, and GLM models are near-frontier, and freely available for anyone to run. While I remain a skeptic about the economics of self-hosting your own Deepseek on a DGX Spark or 17 connected Mac Minis, open models hosted by providers who can benefit from sufficient scale now offer a compelling cost-capabilities tradeoff. And, while I don't expect OpenAI to give all of us the Cursor treatment anytime soon, it's smart to not depend too heavily on on a single provider who may decide to compete with you tomorrow, or decide…

saved by

• Timothy Kostolansky

related reading

• The Myth of unsafe Open Source AIflorianbrand.com
• AI safety advocates: it's time to be more self-reflective about our personal usage of frontier AIconnorsscratchpad.substack.com
• Google "We Have No Moat, And Neither Does OpenAI"semianalysis.com
• Building a harness for open modelsparthsareen.com
• Don’t Outsource Your Thinkingteltam.github.io
• Who’s Afraid of Chinese Models? – Stratechery by Ben Thompsonstratechery.com
• My self-sovereign / local / private / secure LLM setup, April 2026vitalik.eth.limo
• What I learned building an opinionated and minimal coding agentmariozechner.at
• Training Imperativesdan.io
• [2312.06942] AI Control: Improving Safety Despite Intentional Subversionarxiv.org
• Designing agentic loopssimonwillison.net
• Why Open Source AI Will Winvarunshenoy.substack.com