E2B (Execution to Base) is a secure, sandboxed environment for executing code. It is used with UBOS Asset Marketplace for MCP Servers to allow AI models to interact with external data sources and tools.

What is an MCP Server?

MCP (Model Context Protocol) Server acts as a bridge, allowing AI models to access and interact with external data sources and tools.

What kind of LLM providers does E2B support?

E2B supports a variety of LLM providers, including Meta (Llama 3), OpenAI (GPT-4o, O1, GPT-4), Anthropic (Claude 3 Opus), Mistral (Codestral), Groq (Llama 3), Fireworks AI, and Together AI.

Can E2B integrate with AI frameworks?

Yes, E2B integrates with popular AI frameworks like LangChain, LangGraph, Autogen, and Vercel AI SDK.

What are some use cases for E2B with MCP Servers?

Some use cases include AI-powered data analysis, web scraping and data collection, building AI agents with code execution, and integrating with various LLM providers.

✴️ E2B Cookbook

Example code and guides for building with E2B SDK.

Read more about E2B on the E2B website and the official E2B documentation.

Examples

Open-source apps

E2B AI Analyst - analyze your data & create interactive charts
E2B Fragments - prompt different LLMS to generate apps with UI

Hello World guide

TypeScript
Python

Example use cases

Anthropic’s Artifacts UI with AI Code Execution - JavaScript/TypeScript
Upload dataset and analyze it with Llama 3 - Python
Scrape Airbnb and analyze data with Claude 3 Opus and Firecrawl - JavaScript/TypeScript
Visualize website topics with Claude 3.5 Sonnet and Firecrawl - Python

LLM providers

🦙 Meta
- Llama 3.1 405B, 70B or 8B with code interpreter & Together AI
  - Python
  - JavaScript/TypeScript
- Llama 3.1 405B, 70B or 8B with code interpreter & Fireworks
  - Python
- Llama 3 with code interpreter
  - Python
  - JavaScript/TypeScript
OpenAI
- o1 with data analysis and visualization of a csv file
  - Python
  - JavaScript/TypeScript
- GPT-4o with code interpreter and reasoning on image data
  - Python
  - JavaScript/TypeScript
- o1 and GPT-4 with code interpreter performing machine learning on dataset
  - Python
  - JavaScript/TypeScript
Anthropic
- Claude 3 Opus with code interpreter
  - Python
  - JavaScript/TypeScript
- Anthropic’s Artifacts UI with AI Code Execution - JavaScript/TypeScript
- Anthropic models with Firecrawl
  - Visualize website topics - Python
  - Scrape Airbnb and analyze data - JavaScript/TypeScript
Mistral
- Codestral with code interpreter
  - Python
  - JavaScript/TypeScript
Groq
- Llama 3 hosted on Groq + function calling + E2B Code interpreter
  - Python
Fireworks AI
- Firefunction-v2 with code interpreter - Python
- Qwen2.5-Coder-32B-Instruct with code interpreter - Python
Together AI
- Code interpreter usable with Meta Llama 3.1 Instruct Turbo (8B or 70B or 405B), Qwen 2 Instruct (72B), Code Llama Instruct (70B), or DeepSeek Coder Instruct (33B)
  - Python
  - JavaScript/TypeScript

AI frameworks integrations

🦜⛓️ LangChain
- LangChain with Code Interpreter - Python
🦜🕸️ LangGraph
- LangGraph with code interpreter - Python
Autogen
- Autogen with secure sandboxed for code interpreting - Python
▲ Vercel AI SDK
- Vercel AI SDK’s Next.js + AI SDK + Code Interpreter - JavaScript/TypeScript
- Anthropic’s Artifacts UI with Vercel AI SDK - JavaScript/TypeScript