This letter is to everyone in the speedrunning community. It's to everyone who runs, watches, streams, mods, or builds something. I want to lay out exactly where speedrun.io stands right now after the recent speedrun.com terms updates. I've included therun.gg's article below which lays it out extremely well.

therun.gg : The Recent Speedrun.com Terms Changes

Where Things Stand

As of yesterday, I became aware of the term changes on speedrun.com. Before reading through all the community feedback and writing down my thoughts, I first made sure to update the speedrun.io API such that it wasn't in violation of any of the new terms. Here's a quick overview:

  • As of October 2nd, 2026 at 2:58pm EST, all speedrun.io API endpoints (/src, /v1) are returning 503 and will continue to do so until further notice
  • The mirroring of speedrun.com data has been put on hold indefinitely, with data collected up to the end of August 2026
  • The main site has been migrated to archive.speedrun.io and will not be actively updated for the foreseeable future
  • All API release notes and status updates now live exclusively at docs.speedrun.io/notes and docs.speedrun.io/status
  • This follows speedrun.com's October 2026 Terms of Use rewrite, which materially changed the rules speedrun.io was built to operate under

The Current Archive

speedrun.io holds a complete archival snapshot of public speedrunning data collected before this pause under the Creative Commons Attribution-NonCommercial 4.0 International (CC-BY-NC 4.0), ranging from August 2026 back to the project's inception. If you are a verified runner and you want a copy of your own run history as it existed in that archive, I intend to make that available to you directly. This is your data including your own runs. I see it as yours to have, independent of what happens to this project next. I will be posting updates soon outlining how a runner can verify ownership of their data. In the interim, the archive will be stored privately for safe keeping.

The speedrun.io archive was synced from speedrun.com over the course of 7 months, adhering to all API rate limiting rules. The speedrun.io archive consists of:

  • 160,399 Categories
  • 9,730 Developers
  • 1,223 Engines
  • 13 Game Types
  • 51,513 Games
  • 231,479 Levels
  • 236 Platforms
  • 5,529 Publishers
  • 6 Regions
  • 6,115,202 Runs
  • 2,060 Series
  • 708,338 Users
  • 77,333 Variables

Currently On Hold

speedrun.com's Terms of Use were rewritten on September 30, 2026. The new terms include a broad anti-scraping clause with no exemption for API access, competition restrictions, and the revocation of the prior CC-BY-NC 4.0 license that community tools rely on, including this one. This doesn't only affect us; several other long-standing groups were caught off guard by this update.

Continuing to operate with those restrictive changes isn't something I'm willing to do carelessly, so I stopped the archive sync until there's a potential solution. This open letter is a hopeful path to that solution.

Here's a quick list of how things change in the short-term:

  • No current timeline for restoring API access for third-party tools
  • No current plan on republishing data for runners who do not come forward to claim it
  • All development on features outside of archival are indefinitely suspended
  • Not characterizing this as resolved because it isn't

My Promise to You

When I started sketching the initial design of speedrun.io, the first and most important feature was run archival. I actually got a fair amount of pushback that things were fairly fine as is and there were far more pressing things in the community. While I did take the advice and started planning out future features to help the community, I decided to make the archive a fundamental building block of the project. I've experienced several projects, some of them my own, that have suffered at the cost of data rug pulling. In this case, it's a dataset that is built off the time, energy, and sweat of hundreds of thousands of runners, submitted for public record. In my opinion it should remain public record and as a result, I am making the following promise to the community in the hope of gathering support for this vision.

In the near-term, /src, our direct mirror of speedrun.com, is not coming back in its old form. What I'm rebuilding is /v1, our own layer, and only for data a runner has explicitly validated and chosen to share. I do not have current plans to resume the bulk sync of speedrun.com.

What follows is the speedrun.io promise to you. Do note that I will be actively working to rework our entire terms of service and privacy policy, but the existing ones are available on the archival site via the footer.

  • Your current data will be archived, secured and always be made available to you
  • Your data will be made available to you and your community at no cost if you so choose
  • All future data will adhere to these policies
  • This platform is designed with no commercialization in mind
  • No ads, no tracking, no cookies, no donations
  • No barrier, token-based verification
  • Proper rate-limiting and origination checks to curb AI and malicious traffic
  • Full long-term support until at least 2030

The Road Ahead

I still have a lot on my mind. I still think often about our original plans of expanding outside the speedrun.com community and providing the infrastructure for new projects to grow on, runs floating around on YouTube and far more. There's a lot to focus on in the near-term like making verification easy for runners and communities to secure their runs, as well as getting all the missing runs accounted for. In the short-term, the development plan for granular data privacy is the utmost priority and will include the following:

  • The option to restrict any and all public access of your data
  • The option to allow fully public access of your data
  • The option to allow your data to be included in a platform wide data dump for emergency archival migration
  • The option to delete your data with the understanding that it may not be recoverable in the future
  • The ability to download your own data in any way, including scrapers and automation, in accordance with our API rate limiting rules
  • Our own tools to help you with accessing your data easily, posted on our GitHub
  • The /v1 API will be available for your community to build off of, scoped only to data runners have validated and chosen to share

I am hoping to post updates to this letter in the weeks to come, but in the meantime am including a link to the Discord. I will be adding a submission form for additional inquiries in the days to come, specifically for feedback, concerns and willingness to help.

speedrun.io Discord Server

Thank You

I started this project to help the speedrunning community build new and amazing things, but it's now clear that it's time to focus on preservation. I look forward to speaking with anyone from the community who is willing to help.

To get in touch with me before the submission form is live, feel free to email me at [email protected] or ping me on Discord at thekrynn.