GitHub · Agents and tools · Browser and computer use

ThinkFlowLab/system1-agents

Starting from a working agent rather than from an empty loop.

The repository's GitHub card, read on 2026-09-24

What it does

Gives an agent a System One decision model as its brain and ships working examples: browser use, computer use, robotics and games, each of which can be pointed at Jev, a local Laya, or Cua's small model.

The repository's own description: “System 1 decision models (Jev, Laya, Cua-S1) as brain for agents: Browser use, computer use, games and robotics”

How it works

The loop is split so the model only answers bounded questions about what to do next while the harness holds the tools, the state and the safety checks — the same separation the README credits for its speed claims.

The repository, by the numbers

Stars

Licence

Apache-2.0

Stack

Python · Jev / Laya / Cua-S1

Last push

2026-09-23

Read from the GitHub API on 2026-09-24. Stars and the last push move daily — quote them with the date, the way we do.

What we checked

the README, read 2026-09-24
the three supported backends and the four shipped agent types
the licence file

What we did not check. The up-to-6×-faster and 25×-cheaper headline. It compares against a chat model on a task mix the repository does not name, and we did not reproduce it.

Open the repository ↗All builds

The rest of the topic

Jev, the reference →

Start here

Laya, the reference What it is, what we measured running it, and what it is not.
Laya and Jev, compared The two models on the same questions, including the ones Laya loses.

Columns

Built with Laya 18 you are here Ports, servers, agents and games, with what we checked.
What X is saying 22
What people filmed 9
What Hacker News made of it 6
Docs, weights and reading 15

The topic

Every record, by kind of work The same rows cut a second way, across every column.
What self-hosting costs Hardware, memory, and the failure modes of running it yourself.
What the licence allows Apache-2.0 weights, three checkpoints, and the limits that come with them.

By kind of work

Runtime ports 23 Getting the weights onto someone else's hardware: MLX on a Mac, Core ML on the Neural Engine, ggml in C++, candle in Rust, ONNX in Node.
Serving and API 8 Standing the model up as something else can call, usually behind Jev's own wire format.
Agents and tools 6 Typed decisions inside an agent loop: routing, evals, guardrails, memory, and the plumbing around them.
Browser and computer use 2 Deciding the next click, key or field on a real screen.
Games and demos 9 The model playing something you can watch. Most of this model's traffic arrived through one of these.
Benchmarks and evaluation 19 Numbers somebody actually ran — including the ones that say Laya lost.

What it does

How it works

The repository, by the numbers

What we checked

More like this

wdobry/laya-playground

ipenywis/laya-ultrafast

openlayer-ai/jevals