--- title: TinyCodeAgent emoji: 🐣 colorFrom: indigo colorTo: yellow sdk: gradio sdk_version: 5.33.1 app_file: app.py pinned: true license: apache-2.0 short_description: Think in python and execute code to solve a task tags: - agent-demo-track - mcp-server-track - mcp - code-agent --- # TinyCodeAgent 🚀 **TinyCodeAgent: A Self-Building, Self-Debugging Python Agent in the Cloud** 🚀 ## Demo **Watch the demo on YouTube** [![Watch Demo on YouTube](https://img.shields.io/badge/Watch%20Demo-YouTube-red?logo=youtube)](https://youtu.be/CGFSc9Jf5vw) In this demo, watch how **TinyCodeAgent**: 1. **Thinks in Python** * Dynamically generates Python code in the cloud * Instantly executes and tests its own snippets 2. **Builds Its Own Tools** * Creates helper functions on the fly * Integrates new modules to extend its capabilities 3. **Runs & Debugs** * Detects errors, logs tracebacks * Auto-corrects and re-runs until the task is solved 🔍 **Demo Workflow** * **Step 1:** List the most popular Spaces on Hugging Face * **Step 2:** Filter for upcoming Hackathon Spaces without prerequisites * **Step 3:** Display results in a clean, interactive format --- ### Why TinyCodeAgent? * Built on **[TinyAgent](https://github.com/askbudi/tinyagent) **, a lightweight AI agent framework * **Extendable hooks** system for easy customization * Simple event loop ensures stability and fast iteration You can extend it using TinyAgent Hooks system, and use it with Gradio Integration. Your agent could become a MCP Server or tool for another agent. This Project is using **Qwen/Qwen3-235B-A22B** from [Nebius](https://nebius.ai/) thanks to their support. ## Benefits - Stateful python code using Modal Functions (Instead of using a sandbox, Faster and cheaper) - Use TinyAgent Hooks system to extend your agent - Use Gradio Integration to use your agent as a tool for another agent - Log System - Storage - Multi MCP Connections - Support Models through LiteLLM ## Examples - **What are the most popular Spaces on Hugging Face today?** - Please calculate the root mean square error of the following data: - [1, 2, 3, 4, 5] - [1, 2, 3, 4, 5] - How is the weather in Paris? What about the traffic? [Uses dummy functions, for demo purposes] - How is traffic right now in Paris, Montreal, Tokyo and London? ## Credits - [TinyAgent](https://github.com/askbudi/tinyagent) I have built this project as well. - [Modal](https://modal.com/) I use Modal Functions to run python code. - [Gradio](https://gradio.app/) I use Gradio to create a web interface for my agent. - [Cloudpickle](https://github.com/cloudpipe/cloudpickle) I use Cloudpickle to serialize and deserialize python objects. - [SmolAgents](https://github.com/huggingface/smolagents) Use SmolAgents Coding Agent System Prompt as a base for my system prompt.