Deno team joins Cloudflare, ending Deno runtime development after a year

Deno team joins Cloudflare, ending Deno runtime development after a year

In a first-person post on deno.com, the Deno side announced that the entire Deno team is joining Cloudflare. The post frames the move as a way to take forward years of work on simpler server software: how modules are distributed, what security guarantees a JavaScript runtime can give, what a complete toolchain contains, and how easily an app can ship as a standalone executable. Node.js compatibility, the post notes, was an important part of that work because users wanted the Deno improvements while staying plugged into the existing JS ecosystem.

The author says the ambition always went beyond the runtime: compute, storage and communication working together, without every application assembling its own infrastructure. Deno Deploy was a step toward that, but building and operating it showed how much complexity remained underneath. That led to celld, which builds on the Cloudflare Workers programming model and lets developers build distributed applications from the start while keeping the system simple to operate. The author says scaling is built into the programming model rather than being infrastructure each app has to assemble. At Cloudflare the team will combine its work with that of the Workers and Durable Objects teams, with the stated aim of making this programming model the default way to build servers, whether on Cloudflare's network or on the user's own infrastructure.

The post is plain about the cost: joining means choosing where to focus, and future development work goes into this shared platform rather than a separate runtime and hosting service. The author acknowledges this is a consequential change for people who have built on Deno.

The concrete commitments are four. First, the Deno runtime will be supported for another year with monthly releases containing bug fixes and security updates; after that year Deno will end its development of the runtime. It will remain open source, and others who want to continue its development are welcome to. Second, Deno Deploy will keep operating for six months and then shut down, with migration support for paying customers moving to Cloudflare Workers. Third, JSR will continue operating, with its infrastructure moving to Cloudflare. Fourth, Deno will keep supporting rusty_v8 and work toward integrating it into workerd.

The post closes with an AI angle. It says the need for better abstractions is especially acute with AI, and that Durable Objects bring together capabilities particularly useful for agent harnesses: inexpensive, serverless execution, persistent state, WebSockets and a high-level JavaScript interface. That is why celld focuses on Durable Objects. A joint post with Kenton Varda on the Cloudflare blog explains more, and people building agents at scale who want to run them on their own infrastructure are invited to write to ry@cloudflare.com.

Key facts

  • The entire Deno team is joining Cloudflare, per a first-person announcement on deno.com.
  • The Deno runtime gets monthly bug-fix and security releases for another year, after which Deno ends its development; it stays open source and others may continue it.
  • Deno Deploy keeps running for six months and then shuts down; paying customers get migration support to Cloudflare Workers.
  • JSR continues operating with its infrastructure moving to Cloudflare, and Deno will keep supporting rusty_v8 while working toward integrating it into workerd.
  • The team's next project is celld, built on the Cloudflare Workers programming model and focused on Durable Objects, which the post calls particularly useful for agent harnesses.

Why it matters

Deno was built as a challenger to the existing JavaScript runtime model, with its own answers on module distribution, security and toolchain. Its team is now folding that effort into Cloudflare's Workers and Durable Objects platform and will end its own runtime development after a year of maintenance releases. The post presents celld, built on the Workers programming model, as the successor to the Deno to Deno Deploy progression, and says the goal is to make that model the default way to build servers on Cloudflare's network or on a user's own infrastructure.

Who it affects

Teams running the Deno runtime face a one-year support window followed by the end of Deno's own development. Customers of Deno Deploy have six months before the service shuts down, and paying customers get migration support to Cloudflare Workers. JSR keeps operating, with its infrastructure moving to Cloudflare. Developers building AI agents are addressed directly: the post pitches Durable Objects for agent harnesses and invites people building agents at scale to get in touch.

How to use it

There is nothing to install yet; the post is an announcement and a set of commitments. Runtime users get monthly releases with bug fixes and security updates for another year. Paying Deno Deploy customers can move to Cloudflare Workers with migration support from Deno. People who want to keep the runtime alive after that are welcome to continue its development, since it remains open source. Those building agents at scale who want to run them on their own infrastructure are asked to write to ry@cloudflare.com.

How solid is it

This is the primary source: the announcement on deno.com itself, written in the first person on behalf of the Deno team, with specific commitments on timelines. The post says the Deno team is joining Cloudflare; it does not use the word acquisition and gives no deal terms, price or structure. It also gives no date for the announcement or for when the one-year and six-month periods begin. The Cloudflare side is covered only by a pointer to a joint blog post, whose text is not included here.

Risks and caveats

The runtime's development ends after the year of support, so anyone depending on Deno long term needs either to plan a move or to rely on others continuing the open source project; the post names no such group. Deno Deploy shuts down after six months, which is a short runway for production workloads. Migration support is promised for paying customers only, and no pricing or migration details are given. The author states plainly that this is a consequential change for people who built on Deno. The celld plan is described in a few sentences, so how it works in practice is not yet clear from this post.

“We know this is a consequential change for people who have built on Deno.”

— Deno announcement post, deno.com