# Computer use on Jev at about 90ms a decision

DevFeed: [Computer use on Jev at about 90ms a decision](<https://devfeed.tech/articles/computer-use-on-jev-at-about-90ms-a-decision-64568.md>)

Original publisher: [Read original article](<https://jevdaily.com/issues/2026-10-04>)

Author: Jev Daily

Published: 2026-10-04T17:00:00Z

Content type: article

Language: en

Sources: [Jev Daily](<https://devfeed.tech/sources/jev-daily.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [human computer interaction](<https://devfeed.tech/topics/human-computer-interaction.md>)

Tags: [built-with-it](<https://devfeed.tech/tags/built-with-it.md>), [cloudflare](<https://devfeed.tech/tags/cloudflare.md>), [computer-use](<https://devfeed.tech/tags/computer-use.md>), [coreml](<https://devfeed.tech/tags/coreml.md>), [dspy](<https://devfeed.tech/tags/dspy.md>), [langchain](<https://devfeed.tech/tags/langchain.md>), [speed](<https://devfeed.tech/tags/speed.md>), [testing](<https://devfeed.tech/tags/testing.md>), [the-case-against](<https://devfeed.tech/tags/the-case-against.md>)

## AI overview

The issue describes Jev driving a Mac through UI element selection, on-device OCR, and a local CoreML model, with decisions taking about 90 ms and no screenshots or LLM in the loop. It also covers a DSPy model router that can reselect models after tool results, and reports that pg-jev becomes query-speed limited above 50,000 rows.

## Source excerpt

Milind S is driving a Mac with Jev at roughly 90ms per decision, no screenshots and no LLM in the loop. Meanwhile Zachi killed pg-jev after 3 billion tokens...