Cloudflare acquires Deno

So unless someone else picks up development, Deno will no longer be supported.

I hope workerd at least adopts Deno's security mechanisms so it functions as a better sandbox.

One of my favourite projects of 2026 was a configurable LLM harness built around using deno as the runtime. The idea is that the configuration builds a state machine-driven program which is bundled into a binary that can only provide access to the I/O your code explicitly needs. I used it primarily to build interactive programs for colleagues which allow some LLM magic to occur without needing to worry about what they would do with Claude Desktop handling the same data or accessing the same machines and so on. It also allowed for testing local LLMs in deterministic patterns. How do they reason about what to do next when they can evaluate the possible states they can enter, based on the current context? It was really fun. Now I don't really know how I'd rebuild it, knowing I wouldn't use Deno. Maybe that's worth learning anyway.

But I stopped because I saw this coming the moment they changed course and started putting npm compatibility as a priority. Deno’s surface area went from beautifully simple to very bloated. I think they felt the pressure of VC funding and just gave up on rebuilding Node from first principles.

The silver lining is that early Deno was so good that Node copied some of its features. So at least we have a better Node now.

- Cursor -> SpaceX

- Astral/uv -> OpenAI

- Stainless -> Anthropic

- Bun -> Anthropic

- Astro.js -> Cloudflare

- Deno -> Cloudflare

- VoidZero (Vite, etc.) -> Cloudflare

- NuxtLabs -> Vercel

- Hugging Face -> NVIDIA

- ... what else?

What kind of business move is this for Cloudflare? celld is a more complete Cloudflare-at-home runtime than current workerd. What does Cloudflare stand to gain from commodizing Workers?

I'll say that although I'm not really a Cloudflare Workers user, I've been eyeing workerd and celld with interest. The idea of a complete backend in a box appeals to me (see also: PocketBase, Algernon). At the same time, the acquisition means that another company won't acquire Deno for celld.

I've been following celld since it was announced. Bootstrapping both durability and coordination off object storage simplifies so many things for self-hosting. (Yes, ironic that self-hosting has a cloud dependency, but in this case I think justified because S3 has become a widely supported protocol that you can run yourself too).

Will be curious to see the details on exactly how that model makes it into workerd.

Deno's ability to import directly from a package registry or even git repo in a standalone TS script without requiring a package.json or similar 'meta-data' file was actually really nice for shell scripting stuff. AFAIK node.js still can't do anything similar?

of course i’m only talking about deno, the technology not deno, the cloud service.

- there are companies offering serverless functions on their platforms, powered by Deno Deploy. Can this acquisition be an attempt to kill competition?

- Cloudflare now owns a huge chunck of JavaScript ecosystem, like Vercel. They are buying like countries that are arming themselves for war. What is the strategy here?

And worse for Deno - nodejs may not be great but it's good enough.

And may you never have an incumbent competitor that is is "good enough" - it will be your downfall.

Finally, the next step of forking Node is up for grabs:

Node > Deno > Done (anyone?)

[1] https://bunny.net/docs/scripting/

The design and architecture is extremely minimal to the point that all of it can be explained on a single A4 page with a 14pt font including D1 + Durable objects. And I hope that it stays that way.

It has all the primitives that you can wish for to build a software system on top of it be it queues, long running jobs, workflows, pipelines, email handlers, cron jobs and even built in AI models ready for you to be invoked.

ATM - it is extremely cheap, reliable, simpler and more capable than anything out there. Deno itself had very little scope anyway because almost no developer tooling is sellable in this environment even more so post AI. Therefore, it is going to accelerate the Cloudflare platform to be the best in class and hopefully not complex and bloated.

It is a bit hypocritical in a sense. Similar to how I was sad that a local restaurant recently closed down. In the past two years I went there maybe three times total. But I just liked having it there as an option.

Deno had a ton of good ideas but I just never felt confident that it would have the lasting power. Some of the early decisions, like their initial refusal to fully support package.json and the npm eco-system, made me unsure of their suitability as the basis for a business.

But I always wanted them to succeed. In the same way I always wanted Heroku to succeed even though I never used their service.

More options are better. But I guess "use it or lose it" applies. Did I dodge a bullet or contribute to the downfall?

Deno's ability to run TypeScript files - stripping types - was arguably implemented in Node because of Deno. The consolidated tooling approach was a good attempt at solving tooling fragmentation that Node suffers from.

I think for Deno to succeed, it would've had to have been under a non-profit and maybe that can still happen.

I feel bad for anyone that now relies on Deno. And as good as Cloudflare might be for some things, I also feel bad for anyone relying on them.

On the plus side, cloudflare will get access to some fantastic talent, who can hopefully put their effort into building a better web.

https://imgur.com/a/XFAMzOV

No indication cloudflare would pursue that yet that I see.

What's happening to JS platforms? I thought Deno has a much better approach to building a platform.

edit: my reaction was too soon, it seems like they will be explicitly working on making workerd an open source self hostable runtime

I had to evaluate Deno vs. Bun vs. Node for a project two years ago, and chose Node. I think Deno has good ideas but often times these projects cannot reach escape velocity, especially when they're constrained by VC motives.

Whom did they work for before?

Another reason why standards matter. Think twice before you bet on that VC funded horse, folks.

Also worth calling out that if you want to disrupt a major player, you need to be 10x better, not just 2x.

Either way, congrats team. Hope you going on to build something great at Cloudflare.

I know Node is boring but it's not going anywhere.

So what's the story with JSR sticking around then? Does it already see meaningful use among people using Workers, or do they just really want an escape hatch in case we see more long-term issues with GitHub and NPM?

And does it at least mean the stdlib will see continued development?

After Deno did layoffs we saw a consistent decline in announcements and innovation.

Sad day, but unsurprising.

will they continue that work ?

[0]: https://news.ycombinator.com/item?id=49977056

otherwise this is a proper acquisition. t

I know the runtime is in trouble, but i'm not as worried tbh b/c the principles deno championed are going to keep going.

https://blog.cloudflare.com/deno-joins-cloudflare/

TL;DR: Ryan and co. will be merging celld with workerd to create one first-class open source self-hostable runtime for Workers and Durable Objects. In the post I explain in the post why, contrary to what you might think, this is good business for Cloudflare and we're very excited about it.

i always wondered how they were making money

Extremely disappointing...

I've been using deno for years and built some nontrivial services in it. When deno added support for npm packages, the early days were pretty rough. I ran into lots of issues with packages and spent a lot of time reading random github issue threads. It's been pretty smooth sailing for the last year or so though.

At least I can use ai to help migrate off of deno.

https://dbushell.com/2026/03/20/denos-decline-and-layoffs/

I think they ran out of money and joining Cloudflare was their only remaining option.

I'll probably move to Rust for scripting I reckon. It's not very mature but hopefully it won't be too much longer.

Fresh was also the best web framework I've used. Oh well.

I guess I can say that this means Deno won't ever have a much needed Python 3 moment.

(I've trademarked ENDO , feel free to DM me for a license)

Ultimately I only trust Node.js to succeed. Bun being too deep into "shipping anything that increases usage".

Edit: Yes, but it's not named after me, which this fork fixes.

It's going to be the first JavaScript runtime to natively support Super Intelligence APIs.

I'm having an LLM replacing all occurrences of "ai" in every API with "si", in the hopes that OpenSI or SpaceXSI will acquihire me.

> The Deno team is joining Cloudflare to radically simplify self-hosting Workers and Durable Objects so developers can use the same primitives in more places.

Well, that's pretty big news actually.

So Deno is going to be unsupported and will no longer be maintained and will be discontinued.

It looks like "written in Rust" is not enough for a selling point and lost out to the fierce competition against Bun.

Its no longer gonna be supported after 1 year

In that case Deno is not joining cloudflare, it's got eaten.

I invested a lot and use deno everywhere. Can't trust anything these days.

Lets fork it into opendeno. I like to have an all-in-one swiss army knife tool.

Of course, that's not to say the work done on these platforms was wasted. The divide around io.js was meaningful to getting things moving faster and change the governance structure. Deno and bun both showcased things Node could be doing better and some of that was folded back in.

And in spite of all those past experiences, I still find myself using pnpm over npm today. I'm sure npm will continue to evolve and eventually the critical features of pnpm will just be part of core npm. But the gap today is massive enough that pnpm over npm feels critical. The disk usage and performance wins are life-changing for daily work and CI flows.