Agent-S
Agent-S copied to clipboard
Agent S: an open agentic framework that uses computers like a human
Great work! I'm curious if you will release the setup and instructions for Agent-S2 on the AndroidWorld benchmark?
Hello, I’m trying to run Agent S2 in the OSWorld environment and I need your help. I’m running into the following issues: ## 1. Failure to import the Agent S2...
I was configuring Agent S on my system and trying to run **cli_app.py file** using Azure OpenAI credentials. However, I encountered an issue during embedding creation. While debugging the code,...
Currently we support different LLM Providers to execute computer-use workflows locally on macOS Apple Silicon: https://github.com/trycua/cua It'd be interesting to have a new Agent Loop for Agent-S. So far we...
> Hi @xyzhang626 , thank you for the support! > > We use `claude-3-7-sonnet-20250219` as our manager and worker, and use `bytedance-research/UI-TARS-72B-DPO` from HuggingFace as our grounding model. The hyperparams...
- Introduce `mixture_generate_coords` that first tries LLM-based grounding and falls back to OCR on failure for more robust coordinate lookup - Update `assign_coordinates` to use the new mixture method for...
ui tars
ui tars is on open router, can we just use it there? or do we have to run ui tars separate alongside openai/openrouter ? or can we just use open...
Hi, thanks for the great work! I saw that the S2 model shows great results on WindowsAgentArena as well as OSWorld and that the Changelog references gui-agents being easy to...
Simular seems to block popups, which limits the automation I can do. Please allow the user to disable this like in other browsers.