Post-Cutoff

Cua’s 706K-parameter CUA-S1-FORMS beats hosted Jev on form filling

Layton Gott @Layton_GottX

Why it matters

Shows the fast open follow-on to Jev’s ‘System One’ idea (684k views).

Summary

Relays Cua’s open-source CUA-S1-FORMS, a 2.8MB, 706K-parameter ‘System One’ model for computer use that scored 99.7% on Cua’s form-filling eval vs 83.6% for hosted Jev (Cua’s own numbers).

Metrics read 2026-09-29 via api.fxtwitter.com: 683,753 views, 6,769 likes.

Archived text

Yesterday I said Jev would open a ton of doors...

24 hours later, this exists.

Cua built a 2.8MB model that scored 99.7% on their form-filling eval.

Hosted Jev scored 83.6%.

Not to mention it’s FREE and only 706K parameters.

Small enough to run locally with not even 1gb or ram.

Fast enough to make decisions in one pass.

And specialized enough that your agent doesn’t need to call a giant LLM for every tiny action.

Think about what this unlocks.

Every repetitive computer task could eventually get its own tiny specialist:

• forms • CRM updates • data entry • browser actions • document routing • UI decisions

Then one powerful agent just routes work between them.

We are going to see some ridiculous stuff built from this as well.

Quoting @trycua: 1/ Introducing CUA-S1: a family of System One Models, small, specialized, and built for computer use.

Today we’re open-sourcing CUA-S1-FORMS, the first in the family: https://github.com/trycua/cua

Source: x.com/Layton_Gott/status/2101031818413650111Archived 2026-09-29 via fxtwitter (unofficial).Counts: 683,758 views, 6,769 likes, 404 reposts, 93 replies (at fetch time)

Cited in

  1. Model releases 77 days after the cutoff

    TypeSafe AI releases Jev, a ‘System One’ decision model that returns typed probabilities instead of text