simular-ai/Agent-S
Agent S: an open agentic framework that uses computers like a human
Implements vision-based GUI automation through multimodal LLMs paired with a grounding model (UI-TARS) that localizes interactive elements, enabling pixel-level interaction across Windows, macOS, and Linux. The framework achieves human-level performance on OSWorld benchmarks through iterative reasoning and behavior best-of-N sampling, with built-in support for OpenAI, Anthropic, Gemini, and local vLLM backends.
10,117 stars.
Stars
10,117
Forks
1,187
Language
Python
License
Apache-2.0
Category
Last pushed
Feb 21, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/agents/simular-ai/Agent-S"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Featured in
Related agents
elizaOS/eliza
Autonomous agents for everyone
Agent-Field/agentfield
Framework for AI Backend. Build and run AI agents like microservices - scalable, observable, and...
gptme/gptme
Your agent in your terminal, equipped with local tools: writes code, uses the terminal, browses...
SolaceLabs/solace-agent-mesh
An event-driven framework designed to build and orchestrate multi-agent AI systems. It enables...
SmythOS/sre
The SmythOS Runtime Environment (SRE) is an open-source, cloud-native runtime for agentic AI....