I’m using Jev more than Opus 5.5 or GPT-6. Here’s why.

How I AI with Claire Vo · · 26 min

Claire Vo shows how Jev, a cheap classification model, handles sorting, routing, and scoring tasks far better and cheaper than frontier LLMs.

Watch the episode 26 min · YouTube

The gist

  1. Jev returns structured values like scores or yes/no answers instead of text, making it ideal for categorizing or routing large datasets.
    Jev
  2. Vo analyzed 1,700 pull requests by theme in 2 minutes for 9 cents, something that would have cost far more with a frontier model.

What they recommend · 3 picks

Jev Main topic · strong yes

A new AI classification model that takes an input and an output schema and returns probability scores for e...

"More than any other model I've experienced lately, Jev has been the one that has exploded use cases — personal productivity use cases, code use cases, product use cases. It is going to be in the middle of almost everything I build from here on out."
Claire Vo ▶ Play at 0:49

What Claire uses it for: Classification, clustering, and pair-wise grouping of large unstructured datasets including PRs, emails, YouTube comments, and product signals; also used as a fast decision layer alongside smarter LLMs in pipelines.

5 creators recommend Jev, including Riley Brown, Greg Isenberg, Jason Lemkin.

7:39
Codex AI tools

"I'm going to pull up Codex and show you like two or three examples that are super simple to run, incredibly cheap, and incredibly powerful."

Claire Vo ·98 creators recommend it·Claire's AI tools →
8:57
Gemini Flash AI tools

"I use a really cheap model — I use Gemini Flash to then take those clusters and categorize them."

Claire Vo ·74 creators recommend it·Claire's AI tools →

Also mentioned

0:13 Claude · "We have seen a lot of new models be released in the last 5 days. We saw Opus 5.5."
0:13 ChatGPT · "We saw GPT-6 Soul, GPT-6 Luna."
0:21 Grok · "Everybody still loves their Grok bots."
11:23 Claude Code · "All of your Claude Code and Codex sessions are stored locally. And so you can actually run this analysis on..."
16:29 Granola · "These are granola conversations, these are linear tickets — like all the signals in our business about what..."
16:31 Linear · "These are linear tickets — like all the signals in our business about what people are worried about and..."

Ads in this episode

OpenArt Arena. They're paid, so we keep them out of the picks above.

See what How I AI recommends without being paid →
◎

Related episodes

Episodes covering similar topics

JEV: How It Works and What You Can Build
Riley Brown · Sep 18
ai models
Grok 4.5 + Cursor Just Became Unstoppable
Riley Brown · Jul 09
ai models
Claude Fable 5 - is this Mythos model worth the wait?
How I AI · Jun 09
ai models
This New Grok Bot Feature Just Saved Me $500 a Year (Seriously)
Peter Yang · Nov 03
cost optimization
How I AI More from How I AI
Claude Code + Obsidian: My family’s AI brain
Nov 11, 2026
I was a Claude hater. Fable 5.2 changed my mind.
Nov 09, 2026
1,000 days of Duolingo didn't work. So he built his own app with Codex.
Nov 09, 2026
Inside Figma’s PM OS: skills, agents, and Auto XP
Nov 02, 2026
I tested 4 AI agents on 9 real tasks. Here’s who won.
Oct 28, 2026
How Reddit’s designers build ultra-realistic prototypes with Figma + Cursor
Oct 26, 2026

Follow Jev

Free email updates