CoolFace
Apppublic

geraldamasi/agentworld-demo

sourceHugging Faceapache-2.0updated 3mo agoView on Hugging Face
2likes
App README

🌍 AgentWorld Demo — Language World Model Simulator

A Gradio webapp demonstrating the capabilities of Qwen-AgentWorld, a Language World Model that simulates agentic environments. Instead of running real terminals, browsers, or APIs, the model predicts what those environments would return — enabling fast, cheap, and safe agent testing.

What it simulates

DomainDescription
🖥 TerminalLinux shell — predicts command output, tracks filesystem state across turns
🔍 SearchWeb search engine — simulates search results, page extraction, memory store
🔧 MCP / ToolsTool-calling APIs — simulates CRUD operations, API responses, error handling
🌐 WebBrowser — predicts page state after clicks, navigation, form fills

How it works

  1. 1.Pick a domain from the dropdown
  2. 2.Enter an action in JSON format (click "Load Example" to get started)
  3. 3.The model predicts what the environment would return
  4. 4.Multi-turn: the model tracks state across your session (files created, pages visited, etc.)

Model

This Space uses Qwen2.5-7B-Instruct (free HF inference) with the official AgentWorld system prompts from the Qwen-AgentWorld repository.

For full simulation fidelity, deploy the actual Qwen-AgentWorld-35B-A3B model on a paid Inference Endpoint and set INFERENCE_ENDPOINT_URL.

Run locally

bash
pip install -r requirements.txt
python app.py

Credits