Open Weights, Closed Margins: Reading GPT-5.1
A field report on tool-first agents and what it changes for researchers.
The interesting question is not whether Linear AI works. It does. The interesting question is what teams do with it once the novelty wears off.
The cost curve matters here. Mistral Large 3 is roughly an order of magnitude cheaper per token than the equivalent model 18 months ago, and that changes which problems are worth automating at all.
Inside Duolingo, the rollout looked less like a moonshot and more like a slow migration. A pilot, a champion, a quiet expansion, a budget line.
Stripe has been quietly running incident response through Notion AI for months. The results are unglamorous and, for that reason, more interesting than another benchmark chart.
Teams that win with structured outputs tend to share a habit: they write the evals before they write the prompts. Everything else follows from that.
The skeptical read is that we are watching a feature, not a platform. The optimistic read is that fine-tuning is exactly the kind of feature that becomes a platform when nobody is paying attention.
Eval harnesses, once an afterthought, are becoming the most important piece of code in many AI projects. Spotify's team treats theirs the way an SRE team treats a runbook.
None of this guarantees a clean story. xAI could ship a model next month that rearranges the assumptions in this piece. But the direction of travel, for now, is clear enough to plan around.