Fiber: Protocol-Complete, Network-Tested - Development Status, Challenges, and 2026 Outlook

Current Status

Fiber has moved well beyond the purely design or single-channel validation stage.

On the protocol side:

  1. The full channel lifecycle is defined: open, update, close, and dispute
  2. Multi-hop payments across channels are supported at the protocol level
  3. State commitment, forwarding rules, and failure handling are specified with real network behavior in mind

On the engineering side:

  1. A working Rust-based core implementation is in place
  2. Channel state machines and routing-related logic are implemented
  3. The corresponding on-chain contracts and scripts have been implemented to support channel settlement, state verification, and dispute handling

We have been running extensive multi-node, multi-hop network tests over the past year.
They were focused on realistic network scenarios, and they exposed real issues, including:

  1. Fund safety edge cases
  2. Performance bottlenecks under load
  3. Node stability and recovery issues

A significant portion of recent work has been dedicated to fixing these issues and hardening the system.

As a result, Fiber today is not just protocol-complete, but increasingly network-tested.

Positioning Fiber via Lightning Network History

To help set expectations, it’s useful to compare Fiber with the historical development of the Lightning Network.

Lightning went through several phases: starting from a proposal, then making single channels usable, followed by multi-hop routing and real network engineering challenges, and only later reaching broader wallet adoption and ecosystem maturity.

If we map Fiber onto that phases, Fiber today is closer to Lightning’s early network phase.

This means Fiber is no longer just validating basic channel correctness — we are actively validating network behavior under realistic conditions.

At the same time, we are still being careful not to prematurely lock in higher-level assumptions. The key lesson from Lightning remains valid: many long-term constraints come from decisions made too early.

Our goal is to advance network capability while keeping the protocol flexible enough to evolve.

Key Challenges

There are two main challenges we are managing carefully:

  1. Protocol complexity – features like routing, multi-asset support, or watchtowers introduce significant complexity and need to be approached only after the core is stable.

  2. Designing for a more expressive base layer – unlike Bitcoin, CKB is Turing-complete, which gives us much more room to experiment. At the same time, this means we should be careful not to automatically copy assumptions from Bitcoin or Lightning. When designing the protocol and writing the implementation, we want to keep things flexible and avoid locking ourselves into patterns that might be unnecessarily restrictive.

Plan for This Year

This year, our work focuses on three main areas:

  1. Wallet integration
    We are actively pushing integration with JoyID. The goal is to make Fiber usable from a real wallet context, not just as a standalone protocol. This helps validate our API design, UX assumptions, and end-to-end flow under realistic user scenarios.

  2. Business-facing enablement
    In parallel, we will improve documentation and build demos that clearly demonstrate Fiber’s advantages in appropriate scenarios — such as high-frequency payments, low-latency interactions, and off-chain coordination that would be inefficient on-chain.
    These demos are not meant to be marketing material, but concrete references that help wallets, developers, and potential partners understand when and why Fiber should be used.

  3. Liquidity
    Liquidity is a core challenge for both Lightning and Fiber. We will study and reuse proven liquidity solutions from the Lightning Network where they make sense, such as channel management, rebalancing techniques, and liquidity pools. At the same time, we want to explore liquidity designs that are specific to Fiber, rather than simply copying Lightning solutions, and see what works better in this environment.

22 Likes

While we are still deep in the building and refining Fiber, we believe the best way to grow is together!

Whether you’re a seasoned dev or just curious, we’d love to have you on board. Every contribution—no matter how small—makes a huge difference!

Here is how you can dive in:

  • Join the Conversation: Hop into our Fiber Channel! Have a feedback, found a bug, or just have a wild idea? We’re all ears—come say hi.

  • Grab a “Good First Issue”: Ready to get your hands dirty? We’ve curated a list of good-first-issues that are perfect for your first contribution.

  • Level Up our Docs: Our RPC documentation is currently a bit raw. Help us make it more dev-friendly:

    • We’d love to see it presented in a more intuitive way.

    • Interactive Demos would be a game-changer! (Check out LND’s API docs for some cool inspiration).

    • Thinking about bringing in Open-RPC? We’re open to it!

  • Take it for a Spin: Try running a dev or test node with our Quick Start guide. If you hit any snags, don’t keep them to yourself—report them to us! If you have some CKB on mainnet, try to run a Fiber Node for mainnet .

  • Be our “Sanity Checker”: Give our documentation a deep dive. If anything feels fuzzy, confusing, or “off,” please let us know so we can fix it. We’re developing a Fiber Dashboard here https://fiber-dashboard-3aew.vercel.app/, tell us how do you think about it.

9 Likes

Love this approach :ok_hand:

Fiber feels like it’s at exactly the right moment to open things up : Protocol solid, real-world testing underway, now lowering the barrier for contributors and feedback.

Docs, demos, and early UX feedback are often the real unlock for broader adoption. And this kind of work feels especially important for making CKB-based infrastructure approachable beyond the core dev crowd.

Looking forward to seeing how the ecosystem builds around it.

4 Likes

We’re gonna take it for a spin

3 Likes

Can I integrate Fiber into web browser through WASM tech now? and I wonder if Fiber has featured the intranet penetration for the interaction among all of Fiber nodes running inside the browser?

2 Likes
  • For the first question: Yes. We have fiber wasm now, and it’s published on npm @nervosnetwork/fiber-js. It’s based on wasm and runs in browser. There is also a demo bases on fiber-js fiber-wasm-demo.vercel.app

  • For the second question: No. We have no support for intranet penetration now

2 Likes

cool for your clarification, but without intranet penetration, how does Fiber interact with other ones?

  • For non-wasm nodes, they are required to be on Internet with public access
  • For wasm nodes, they can be in the intranet, and only keeps outbound connections

Sorry, what is about outbound connections? For instance, between two different wasm nodes, they can communicate only in case of connecting a same public non-wasn node?

A `outbound connection` of node A means A has a connection to another node on Internet.

For your question, yes. Wasm nodes can’t have inbound connections.

However, there isn’t a restriction that different wasm nodes should connect a restricted SAME public node before starting their mutual communication, right?

They can connect any desired public node since all Fiber nodes consist of an entire network, in which, any two nodes can establish connection at any time.

Correct me if I’m wrong.

Yes, you are right

Hey @quake, @chenyukang & @Officeyutong, I would like to thank you for creating this report, being responsive to Community questions and creating a Fiber channel on Discord. Thank you so much for stepping up and keeping the CKB vision alive :fire:

As for me, I’m personally curious about cross-chain channels.

For example, with the upcoming Rosen Bridge integration, we’ll get the same asset represented on multiple chains, including Nervos L1. Ideally this should facilitate cross-chain channels without having to depend on oracles and market makers.

So I was wondering:

  • What’s your position on cross-chain channels?
  • Do you have any plans/idea on that direction?
  • Any kind of comment you’d like to make?

Love & Peace, Phroi

2 Likes

on this topic check out Stable Channels

1 Like

Hey @matt_ckb, thank you for the link, pretty interesting Contract idea, at high level:

  • A form of Perpetual Futures Contracts
  • Non Tradeable Contracts, strictly P2P
  • Coin-Margined Futures, so funded by native assets
  • Targets an external Price Feeds
  • Settled via continuous LN payments

Downsides:

  • Price feed: it still uses off-chain oracles, which can be manipulated directly or indirectly.
  • Cumbersome: there is a high difficulty for end-users in running the system, especially considering the active management needed.
  • Intrinsic Complexity: it’s already difficult and error prone when done by CEX, I can’t really imagine this P2P DEX version ever becoming 100% safe to use.
  • Limited Upside: that we like it or not, centralized stablecoins and CEX are already fulfilling this user need at an almost zero user cost, pretty difficult to beat that with this system.

In a way, this is the exact opposite of what I was hinting in the previous comment and more of a continuation of our previous conversations :hugs:

Thank you for taking your time to find this interesting Channels application,
Phroi

3 Likes