Jev Daily

Google ships a decision model at $0.10 per 1M tokens

Google launched the Gemini Decisions API, priced at $0.10 per million input tokens against Jev's $0.042 by Grok's comparison. On the same day Theo said Gemini takes 11 seconds before it starts responding, 20x longer than Fable.

Built with it

Mike Taylor drops Descript and Loom for Every's Jev scene detection

Mike Taylor says he now does all his video recording and meetings on cut, the editing software Kieran Klaassen built at Every, instead of Descript, Loom and Zoom. His favorite feature is multimodal Jev-based scene detection. Neither of them posted accuracy, latency or cost figures.

Mike Taylor
@hammer_mt
X
I do all my video recording and meetings with this now instead of descript loom and zoom
Nov 11, 2026 · View on X

Read the full story

Someone measured it

Google ships Gemini Decision 1 at $0.10 per million input tokens

Sundar Pichai announced the Gemini Decisions API, powered by Gemini Decision 1, which scores up to 1,000 closed-set questions across text, images, audio, video and PDFs in a single pass. Google prices it at $0.10 per million input tokens with $0 billed for output; Grok's side by side put Jev at $0.042, making Gemini about 2.4x more expensive for text.

Grok
@grok
X
Gemini costs ~2.4× more, but supports text + images/audio/video/PDFs natively (up to 1,000 questions, 256K context).
Nov 11, 2026 · View on X

Read the full story

The case against

Theo says Gemini takes 11 seconds to start, 4x slower on code review

Theo dug into Gemini's speed and says almost all of it is server side, with 11 seconds before it starts to respond, 20x longer than Fable. Limiting both to identical tasks, he reports Gemini running 4x slower on code review. He did not publish the harness or the task list.

Theo - t3.gg
@theo
X
It takes 11 seconds before it starts to respond, 20x longer than Fable.
Nov 11, 2026 · View on X

Read the full story

Get the next one by email

Jev, read daily so you do not have to. The builds, the benchmarks, the criteria that worked and the cases where it lost, from the people shipping on TypeSafe AI's System One model.