Introducing Together Code Sandbox & Together Code Interpreter: SOTA code execution for AI

Together AI · 1 year ago

Together AI launched Together Code Sandbox and Together Code Interpreter, two products designed to enable secure code execution for LLM-generated code. Together Code Sandbox offers customizable virtual machine environments with VM startup times of 500 milliseconds (P95) and pricing at $0.0446 per vCPU per hour, while Together Code Interpreter provides session-based Python code execution via API at $0.03 per 60-minute session. The products address delays and infrastructure complexity developers face when running AI-generated code, with existing users like HeroUI reducing development time from five months to two weeks.