AgentENV: Distributed Platform for Large-Scale Agent Environments

Repositorykvcache-ai/AgentENVTool

AgentENV (AENV) is an open-source Rust-based platform for running isolated agent environments at scale, notably powering agentic RL training for Kimi K3.

Key points

  1. Uses Firecracker microVMs and OverlayBD to stream OCI-compatible images on demand with bounded local caching.
  2. Delivers rapid lifecycle transitions: snapshot-backed boots and resumes under 50 ms, pause operations under 100 ms.
  3. Features native incremental memory and filesystem snapshots, enabling running instances to fork into parallel sandboxes.
  4. Achieves up to a 9.6x memory overcommit ratio in production using ublk I/O and memory ballooning.
  5. Requires modern host infrastructure with Linux kernel 6.8 or higher and KVM access.

Why it matters

Reinforcement learning for AI agents demands fast, ephemeral, and strictly isolated sandboxes across diverse workloads, which traditional virtual machines or standard container setups struggle to provide economically.

Analysis

AgentENV addresses density and startup bottlenecks by pairing lightweight Firecracker microVMs with on-demand block-level image streaming. It is already battle-tested in Moonshot AI's Kimi K3 training pipeline at massive scale. However, operators should note current operational limits, such as unencrypted API endpoints and bleeding-edge kernel requirements.

Mentioned links

Found in

Summary generated by AI (Gemini 3.8 Flash, 27 September 2026). It may contain errors – the original source is authoritative.

Related entries

Similar in content, found by the AI search.

  • Repository:agent-substrate/substrate

    Agent Substrate: Zero-Trust Sandboxing for Autonomous Agents on Kubernetes

    Agent Substrate is an open-source Go runtime designed to run millions of agent workloads securely and efficiently on Kubernetes. It isolates untrusted code using microVMs or gVisor and snapshots idle agents to save resources.

    5.4kstarsGo

    AI & LLMs· Tool

  • Repository:NVIDIA/OpenShell

    NVIDIA OpenShell: Isolated Runtime for Autonomous AI Agents

    NVIDIA OpenShell is an open-source runtime written in Rust designed to run autonomous AI agents in secure sandboxes. It enforces declarative policies at the kernel level and guards access to network routes and credentials.

    16.6kstarsRust

    AI & LLMs· Tool

  • Repository:google/ax

    Google AX: Declarative Orchestration Runtime for AI Agents

    Google has open-sourced AX, a declarative orchestration runtime designed to manage autonomous AI agent workloads. Built on Kubernetes and Agent Substrate, it handles sandboxed execution, workspace pre-wiring, and network fencing at scale.

    6.2kstarsGo

    AI & LLMs· Tool

  • Repository:LMCache/LMCache

    LMCache: Scalable KV Cache Management Layer for LLM Inference

    LMCache is an Apache-2.0 open-source caching layer designed to accelerate LLM serving by optimising and sharing Key-Value caches across processes and network nodes on Nvidia and AMD hardware.

    11.7kstarsPython

    AI & LLMs· Tool

  • Repository:CopilotKit/OpenBot

    OpenBot: Open-Source Platform for Isolated and Governed AI Coworkers

    OpenBot by CopilotKit is a self-hosted platform providing AI agents as coworkers. Each agent operates within its own dedicated computer environment, featuring a real browser, scoped file access, and explicit tool permissions.

    4.3kstarsTypeScript

    AI & LLMs· Tool

Back to the knowledge base

Let's talk about your project

Get in touch

Locations

  • Mattersburg
    Johann Nepomuk Bergerstraße 7/2/14
    7210 Mattersburg, Austria
  • Vienna
    Ungargasse 64-66/3/404
    1030 Wien, Austria

Parts of this content were created with the assistance of AI.