Raycast built its Ray 1 models by fine-tuning GPT-4o and mini
Thomas Paul Mann · Raycast: Your AI Automation Assistant · Feb 26, 2025 · at 8:15
Thomas Paul Mann is the CEO of Raycast. He explains the technical underlying architecture of the newly announced Ray 1 models powering AI Extensions.
0:00 / 0:13exact quote · 13.8s
▶ Watch the full episode on YouTube → 720p mp4 · rendered on demand · StarZero watermark
“And so we looked into all the various models we had and then we picked, at the moment, it's gbd-for-o and gbd-for-o-mini, Which we basically did a fine tune to really optimize for our use case, and then basically shipping that in the app as Ray one and Ray one mini.”
quote is from the automated transcript, cleaned for reading: filler sounds and stutters are removed, nothing is rephrased. names can be misheard (the analysis reads context, assessments check outside sources). how →
More from Thomas Paul Mann
OpenAI models remain the industry best for chained function calling
“And so we find like the open AI models function calling wise for our use case. So for the best ones and like, yeah, basically verifying the others. We had a beta group and testing different models. So basically from all providers and yeah, find basically the b…”
Mann: AI evaluations for tool calling are super unsolved across industry
“Nobody talks about how to do evals with tool calls. It's, like, a super unsolved thing.”
Raycast CEO: Incumbent operating systems are too mature to innovate
“But I can't really see that coming from the big operating systems. They're like so mature. They're like nothing moves anymore.”
Mann: OpenAI-Compatible APIs Still Have Subtle Behavioral Differences
“Like, even though all models say they support open AI compatible APIs, they're always in nuances which are slightly different and then put you off when you see it for the first time.”