NOW LET US – AI RAG SaaS Studio TP.HCM
NOW LET US
Digital Product Studio
Back to news
DEV-TOOLS...2 min read

Deno Desktop

Share
NOW LET US Article – Deno Desktop

Deno Desktop, introduced in Deno v2.9.0 (canary), compiles Deno projects into self-contained desktop applications with small binaries and native performance.

deno desktop

turns a Deno project (anything from a single TypeScript file to a Next.js app) into a self-contained desktop application. The output is a redistributable binary that bundles your code, the Deno runtime, and a web rendering engine into one bundle per platform.

deno desktop

ships in Deno v2.9.0 and is not in a stable release yet. To try it now, run deno upgrade canary

to install the canary

build. The command, configuration keys, and TypeScript APIs may still change before the feature is stable.

Why deno desktop

Jump to heading

Web technology is the most widely-known UI toolkit in the world. Desktop apps built on web stacks (Electron, Tauri, Electrobun) take advantage of that, but each has tradeoffs you have to live with: huge binaries, missing platform support, no JavaScript ecosystem, no built-in update story, no framework integration.

deno desktop

is opinionated about those tradeoffs:

**Small by default, full Node compatibility.**The default WebView backend uses the operating system's own webview for small binaries, and you still have the entire npm ecosystem available through Deno's Node compat layer. Opt into the bundled Chromium (CEF) backend when you need identical rendering across macOS, Windows, and Linux.**Framework auto-detection.**Pointdeno desktop

at a Next.js, Astro, Fresh, Remix, Nuxt, SvelteKit, SolidStart, TanStack Start, or Vite SSR project and it runs: the production server in release mode, the dev server with hot reload under--hmr

. No code changes are required to take an existing web project to the desktop.**In-process bindings instead of IPC.**Backend and UI communication goes through in-process channels, not socket-based IPC. Values are still encoded as they cross the call boundary, but there is no cross-process round-trip between your Deno code and the webview.**Cross-compile from one machine.**The same machine can build for macOS, Windows, and Linux. Backends are downloaded as needed, not built locally.**Built-in binary-diff auto-update.**Ship a singlelatest.json

manifest and bsdiff patches; the runtime polls, applies, and rolls back automatically on failed launches.

Hello, desktop Jump to heading

Create a one-file desktop app:

Deno.serve(() =>
new Response("<h1>Hello, desktop</h1>", {
headers: { "content-type": "text/html" },
})
);
deno desktop main.ts

The compiled binary opens a window pointed at a local HTTP server bound to your Deno.serve()

handler. Run it directly:

./main # macOS / Linux
.\main.exe # Windows

Deno.serve()

automatically binds to the address the webview navigates to, so you do not need to pass a port or hostname. See HTTP serving for details.

What's in this section Jump to heading

  • Configuration: the desktop

block indeno.json

. - Backends: CEF, webview, raw; how to choose.

  • HTTP serving: Deno.serve()

integration and the serving model. - Frameworks: Next.js, Astro, Fresh, Remix, Nuxt, SvelteKit, and others.

  • Windows: Deno.BrowserWindow

lifecycle, multiple windows, events. - Bindings: calling Deno code from the webview via bindings.<name>()

. - Menus: application and context menus.

  • Tray and dock: system status icons and the macOS dock.
  • Dialogs: prompt()

,alert()

,confirm()

as native popups. - Notifications: native OS notifications via the Web Notification

API. - Hot module replacement: --hmr

for framework and non-framework apps. - DevTools: unified DevTools attached to both the Deno runtime and the webview.

  • Auto-update: Deno.autoUpdate()

, manifests, bsdiff, rollback. - Error reporting: capturing uncaught exceptions and panics.

  • Distribution: cross-compilation, output formats, installers.
  • Comparison: how deno desktop

relates to Electron, Tauri, Electrobun, Dioxus. deno desktop

CLI reference: the command, its flags, and thedeno.json

desktop

schema.

© 2026 Now Let Us. All rights reserved.

Source: Hacker News

Advertisement
Ad slot ready: 5887729102

More in this category

NOW LET US Related – Codex logging bug may write TBs to local SSDs

dev-tools

Codex logging bug may write TBs to local SSDs

A logging bug in Codex continuously writes massive amounts of data to local SQLite databases, potentially reaching 640 TB per year. This high write volume can quickly exhaust the write endurance (TBW) of consumer SSDs, destroying them in less than a year.

NOW LET US Related – GLM 5.2 vs. Opus

dev-tools

GLM 5.2 vs. Opus

A head-to-head comparison between the new open-weights model GLM-5.2 and Claude Opus 4.8 in building a 3D platformer from scratch using raw WebGL.

NOW LET US Related – Sakana Fugu

dev-tools

Sakana Fugu

Fugu-Ultra, a new AI agent system, has demonstrated superior performance over leading frontier models across six complex real-world benchmarks. From autonomously optimizing GPT training recipes and decoding classical Japanese manuscripts to playing blindfold chess and stock trading, Fugu-Ultra showcases the immense potential of orchestrated AI agents.

NOW LET US Related – Good results fine tuning a local LLM like Qwen 3:0.6B to categorize questions

dev-tools

Good results fine tuning a local LLM like Qwen 3:0.6B to categorize questions

A detailed guide on fine-tuning a tiny local LLM (Qwen 3:0.6B) to categorize household questions for metadata-aware RAG, boosting accuracy from 10% to 92% using Unsloth and opaque ID mapping.

NOW LET US Related – Did my old job only exist because of fraud?

dev-tools

Did my old job only exist because of fraud?

A software engineer reflects on a shocking realization that his early-career startup job, which relocated him to the US, might have only existed as a vehicle for a multi-million dollar venture capital fraud exposed by the SEC.

NOW LET US Related – Apertus – Open Foundation Model for Sovereign AI

dev-tools

Apertus – Open Foundation Model for Sovereign AI

Apertus has officially released Apertus Mini, a set of 16 small language models demonstrating advanced distillation and quantization techniques. Designed to meet EU AI Act requirements and support over 1,000 languages, Apertus aims to redefine open-source and sovereign AI.

EXPLORE TOPICS

Discover All Categories

Deep dive into the specific technology sectors that matter most to you.