X releases its For You ranking code on GitHub alongside an account labels tool
X has open-sourced its For You timeline ranking algorithm on GitHub under the Apache v2 license, including signal weights and core ranking logic, while also launching a new account labels transparency tool for frequent posters.
@X has published the source code powering its For You timeline on GitHub, marking one of the most significant transparency moves by a major social platform. The company made the source code for the For You timeline, the default feed that opens when you launch the app, available on GitHub under the Apache v2 license. The release expands on earlier open-source efforts by adding the model configuration, filters, and core ranking system details, including the parameters used to weight different signals, which are key to understanding which posts are actually displayed.
A Much Larger Codebase
The update makes the codebase roughly 10 to 15 times larger than it was before. The repository contains the core recommendation system powering the For You feed, combining in-network content from accounts you follow with out-of-network content discovered through machine learning, ranking everything using a Grok-based transformer model.
Alongside the code release, @X is rolling out an account labels tool that gives users more visibility into how their accounts are classified. The feature will allow frequent posters to download a file listing any labels applied to their account, starting as a pilot for accounts at least one year old. Rule enforcement driven by @grok remains private and is not included in the public release.
Freedom of Speech, Not Reach
The release comes as X continues to operate under its Freedom of Speech, Not Reach enforcement philosophy, which restricts the reach of posts that violate its policies by making that content less discoverable. X has expanded the application of this enforcement action beyond hateful conduct to also cover abusive behavior and violent speech. Labels applied under this policy display which rule the post potentially violates to both the author and other users, and labeled posts are made less discoverable on the platform.
Compared to a standard post, these labels reduce the reach of violative content by 81 percent. Posts hit by a restricted-reach action carry a visible label, giving authors notice that their distribution has been limited. Authors are able to submit feedback on the label if they believe their post's visibility was incorrectly limited.
Together, the algorithm release and the labels tool represent @X's clearest effort yet to give users a direct line of sight into how content is ranked and moderated on the platform.
Sources:
TechCrunch: X open sources its ranking algorithm
X Blog: Freedom of Speech, Not Reach enforcement philosophy
GitHub: xai-org/x-algorithm repository
Latest News
Read More...
Author
Crypto RichRich has been researching cryptocurrency and blockchain technology for eight years and has served as a senior analyst at BSCN since its founding in 2020. He focuses on fundamental analysis of early-stage crypto projects and tokens and has published in-depth research reports on over 200 emerging protocols. Rich also writes about broader technology and scientific trends and maintains active involvement in the crypto community through X/Twitter Spaces, and leading industry events.













