Monty v1 open-sourced: Python sandbox starts in 1ms, runs 10k scripts in 674ms
samuelcolvin · x · 2026-09-25
Pydantic author Samuel Colvin released Monty v1, an open-source Python sandbox that starts in 1 millisecond instead of 1.5s, targeting the biggest pain point of letting AI agents write code.
Key points:
- Benchmarked 10k sandboxed scripts in 674ms — a task he says would take a cloud sandbox over 3 hours.
- Supports long-lived sessions: Monty can be dumped and resumed at any external function call, useful for stateful agents.
- Installable via PyPI, npm, or Crates; a hosted platform is coming and design partners are being recruited.
The same post also surfaces OmniCoreAgent, an open Python agent runtime for production with tools, MCP, memory, workspace, telemetry, subagents, background tasks, and OmniServe APIs.
Related event: Pydantic creator open-sources Monty, a 1ms-startup Python sandbox(3 posts)→
More from coding & agent
- Runway MCP Lets Any AI Agent Generate Video Directly — tlakomy · 2026-09-25
- Hexis: open-source Postgres-native cognitive architecture giving LLMs memory and identity — GeretsOliver · 2026-09-25
- AgentArk, distilling multi-agent debate into a single LLM, accepted to NeurIPS 2026 — jindong_wang92 · 2026-09-25
- Ben Tossell Built an 87-Device Retro Timeline Site in a Morning with Codex and Subagents — Ben's Bites · 2026-09-25
- video-shotcraft author: Opus 5.5 makes a full code-generated video with music in one prompt — caolonghua · 2026-09-25
- Stability Matrix hang on Linux traced to 143 Windows fonts breaking Fontconfig — Signal_Care6558 · 2026-09-25