By Michelle Lee —

IPFS Data Utilities Grants, Year 1: Primitives in Motion

IPFS was built on a simple premise: content should be identified and verified by what it is, not where it lives. For most of its history, these ideas lived at the edges of the web: elegant and principled, but relatively niche. Then the internet filled up with AI-generated media and algorithmic capture, and the problems IPFS was built to solve became impossible to ignore. Last year, we introduced a new protocol suite and grants program to meet the moment. Here's what we funded in Year 1, what shipped, and what we learned.

Background: From Full Nodes to Sharp Primitives

In 2022, the IPFS Implementations Grants launched to support the development and diversification of IPFS — backing projects to build full IPFS nodes for new languages and environments like mobile and Electron, ground-up rewrites of core libraries, and browser modernization. Highlights from this period took about 3 years to crystallize, including Iroh’s new peer-to-peer networking layer for devices, and the recent announcements of Ed25519 and protocol handler support in Chrome.

Meanwhile, we saw a new dynamic emerging in the IPFS developer community. People weren't asking for big new full node implementations. They wanted small tools that did one thing well, and let them get back to work. When they couldn't find them in the existing IPFS stack, they built their own. During 2024-2025, nearly a dozen small, sharp implementations like atcute, libipld, and more emerged, many from the fast-moving atproto community.

We took that as a signal. In early 2025, we launched DASL, a lean and modular subset of IPFS protocols. Alongside it, we launched the Data Utilities grants, sized to match. Three cycles, 9 grants, and $200,000 in, here's what we've learned.

How the Program Works

Our Data Utilities Grants run $5,000 to $25,000, with timelines of 1-3 months. This is typically enough to support one building block done well (including research, design, implementation, docs, user testing, and so on). For protocol or interoperability work, we also make sure the grant scopes cover the human costs of cross-project coordination. 

Each cycle has a focused set of RFPs that identify a problem space without prescribing a solution. We prioritize teams that are already building for (and directly accountable to) existing users. The strongest proposals are ones where multiple projects are lined up and ready to test and ship the resulting work.

Grants Don’t Work Alone

In parallel, our IPFS Foundation team is pushing modularity at the protocol level — most visibly through DASL (Data-Addressed Structures & Links), which separates content addressing, data transfer, and routing so each can evolve independently. Several of the grants awarded harden or extend DASL to adapt it for the grantee’s domain-specific needs.

The third leg is the monthly CID Congress virtual series, which convenes protocol authors, implementers, and developer-users in a friendly and collaborative setting. Many RFPs in this program came directly from needs surfaced there. It's our best instrument for ensuring we're funding things developers actually want and need.

What Shipped

  • The DASL test suite (Hypha) immediately found several issues across implementations and worked with maintainers to fix them. These included CBOR integer ranges, UTF-8 string validation, and more.
  • go-dasl (Hypha) was merged into Bluesky, streamlining dependencies lists and making AT Protocol development more approachable for new devs.
  • dasl (rust) (Number Zero) is currently in evaluation by a project with many millions of users and real-time needs.
  • The BDASL spec (Number Zero) closes a longtime gap for anyone working with large or time-sensitive data: range requests and streaming verification. This work fed immediately into MUXL and S2PA.
  • (Coming soon) MUXL and S2PA (Stream.place) are extending DASL to bring signed, content-addressable primitives to video.
  • (Coming soon) NIIFTY2 (Mnemoscene) is extending DASL and metadata publishing to IIIF-based cultural heritage collections.
  • And more!

We’re finding that small and modular doesn't just apply to software design. A modular protocol ecosystem is best grown through modular work: scoped, interoperable, and composable over time. 

That said, small isn't always right. Our longer-running Implementations grants continue to support transformative work where the arc of change is genuinely long. (Browsers, we're looking at you.)

What's Next

For developers, we're currently in the final round of review for Spring 2026 applications focused on DASL fuzz testing and extending content addressing to new domains. The next RFP round will open in late 2026.

Through 2026 and beyond, our IPFS Data Utilities grants will continue to support critical building blocks for a world of AI-generated media, fragile platforms, and users who deserve algorithmic choice and data autonomy. 

This work is stronger together. If you believe the open web is worth investing in, we'd love to talk.