deeplethe/forkd Rust

🌐 forkd

Fork() for AI agent microVMs. Spawn 100 children in ~100ms from a warm parent; BRANCH a live VM in ~150ms. KVM-isolated, snapshot CoW.

🌐 用于AI代理microVM的Fork ()。在大约100毫秒内从温暖的父母生成100个孩子;在大约150毫秒内分支实时虚拟机。KVM隔离,快照CoW。

⭐ 691Stars
🍴 47Forks
🐛 8Issues
🕐 2026-05-25Updated
ai-agentscopy-on-writekvmmicrovmrustsandboxsnapshot
Project Insight

Why It Is Worth Watching

From the perspective of project description, tags, and public data, it is worth looking at deeplethe/forkd first. It solves problems around the needs of AI applications, agents, or large model toolchains. Related tags include ai-agents, copy-on-write, kvm, and microvm, mainly using Rust. The most direct message in the project description is: “Fork () for AI agent microVMs. Spawn 100 children in ~ 100ms from a warm parent; branch a live VM in ~ 150ms. KV…”. I will put it on the observation list, not because of the number of Stars itself, but because of its direction and recent popularity: about 52 Stars were added in the past day, recently updated on 2026-05-25, 691 Stars and 47 Fork; the most recent code update was on 2026-05-25, this statistic added about 52 Stars relative to the historical benchmark. If it can continue to grow in the next few days, at the same time, Issue, PR, and submission records are also relatively active, indicating that it may not only be a wave of forwarded for a short time, but actually stepped on the recent needs of developers. It is more suitable for AI product developers, independent developers and large model landing teams, especially when technology selection is being done

📈 Trend History (4 points)