Agent-S
Agent S: an open agentic framework that uses computers like a human
Benchmark-leading computer-use automation driven by screenshots.
You run multiple monitors — the agent is designed for single-monitor screens.
About Agent-S
Welcome to Agent S, an open-source framework designed to enable autonomous interaction with computers through Agent-Computer Interface. Our mission is to build intelligent GUI agents that can learn from past experiences and perform complex tasks autonomously on your computer.
Whether you're interested in AI, automation, or contributing to cutting-edge agent-based systems, we're excited to have you here!
On OSWorld, Agent S3 alone reaches 66% in the 100-step setting, already exceeding the previous state of the art of 63.4% (GTA1 w/ GPT-5). With the addition of Behavior Best-of-N, performance climbs even higher to 72.6%, surpassing human-level performance on OSWorld (~72%)!
Agent-S is an open-source project written primarily in Python, with 12k stars on GitHub. It was last updated in August 2026.
pip install gui-agentsAgent-S vs. the alternatives
All 19 alternatives →| Record | Stars | Pricing | ||
|---|---|---|---|---|
| Agent-Sthis listing | 12k | Python | Apache-2.0 | Open source |
| UI-TARS-desktop | 39k | TypeScript | Apache-2.0 | Open source |
| page-agent | 29k | TypeScript | MIT | Open source |
| skyvern | 23k | Python | AGPL-3.0 | Open source |
| midscene | 15k | TypeScript | MIT | Open source |
| nanobrowser | 14k | TypeScript | Apache-2.0 | Open source |
