Skip to main content

Jev, the AI at a fiftieth of OpenAI's price: is it any good?

Jamie Watters

Operational resilience and AI delivery practitioner.

Published: 10 October 2026•Watch on YouTube → (opens in a new tab)
Jev, the AI at a fiftieth of OpenAI's price: is it any good?

Watch on YouTube →

Jev is a new AI model that costs a fiftieth of the price of OpenAI's GPT-5.4. On my first test it beat GPT-5.4 by 11 points, and I told the internet. Two days later I ran the same test again, and the win was gone.

There's a lot of hype in AI right now, and most of it arrives as a chart where one model beats another. This video shows how my own chart fooled me, and the one question that cuts through every chart like it: can I see every run?

The verdict on Jev: on this job I couldn't tell it apart from GPT-5.4, at a fiftieth of the price. That's a good story. It's just not the one the hype tells.

In this video:

  • The test, and the result I published
  • What five runs showed instead
  • How my own marking picked the winner
  • The one question to ask of any AI comparison chart
  • The real news about Jev

What Jev is and when to use it (the full explainer): https://jamiewatters.work/journey/jev-explained

How to test an AI model yourself, with a full checklist: https://jamiewatters.work/journey/how-to-test-an-ai-model

Every claim, every run and the marking, open to check: https://github.com/TheWayWithin/jev-bench

More of my writing on AI, tested and checked: https://jamiewatters.work/journey

No paywall, no sponsors. If this saved you some time, you can buy me a coffee.

☕ Buy me a coffee

Get the next one in your inbox

I build with AI in the open and write up what held and what didn't. Real numbers, the failures before the wins.

Share this post