X Algorithm Weights, Right Now

Every ranking weight, parsed directly from param.rs. These multiply predicted probabilities — not raw counts.

Auto-updated when the repo changes — last updated 2026-08-15live

Positive Weights (16)

NameValueDescription
BidirectionalFollowReplyWeightBoost
15Bonus on reply weight for mutual-follow original posts
ClickWeight
0.4Predicted probability you'll click into the post
ContDwellTimeWeight
0.004Continuous dwell time contribution
FavoriteWeight
0.5Predicted probability you'll like the post
FollowAuthorWeight
4Predicted probability you'll follow the author
OpenLinkWeight
0.2Predicted probability you'll open a link
PhotoExpandWeight
0.05Predicted probability you'll expand a photo
PostUnexploredWeight
0.02Exploration: posts outside your usual patterns
QuoteWeight
5Predicted probability you'll quote
ReplyWeight
5Predicted probability you'll reply
RetweetWeight
1Predicted probability you'll retweet
ShareViaCopyLinkWeight
20Predicted probability you'll copy the link
ShareViaDmWeight
5Predicted probability you'll share via DM
ShareWeight
2Predicted probability you'll share
VideoOpenWeight
0.05Predicted probability you'll open a video
VqvWeight
0.05Video quality view weight

Negative Weights (5)

NameValueDescription
BlockAuthorWeight
-31.2Predicted probability you'll block the author
MuteAuthorWeight
-58.8Predicted probability you'll mute the author
NotDwelledWeight
-0.02Penalty for posts you didn't dwell on
NotInterestedWeight
-43.2Predicted probability you'll mark not interested
ReportWeight
-234Predicted probability you'll report the post

Adjustments & Factors (8)

NameValueDescription
AuthorDiversityDecay
0.5Decay factor for same-author posts in one session
AuthorDiversityFloor
0.25Floor for author diversity decay
ColdStartFollowerCap
1000New-author boost applies below this follower count
ColdStartImpressionThreshold
1000New authors boosted until this many impressions
ColdStartSlotMax
16Maximum feed position for new-author boost
ColdStartSlotMin
15Minimum feed position for new-author boost
OonWeightFactor
0.75Discount applied to out-of-network posts
TopicOonWeightFactor
0.5OON discount when topic filtering is active

Weights Misconceptions

Do these weights multiply raw engagement counts?

No. The repo is explicit: weights multiply the predicted probability of each action, not raw counts. A ReportWeight of -234 does not mean one report cancels 468 likes. It scales how much the predicted likelihood of a report affects the ranking score.

Why is the Report weight so negative (-234)?

The baseline probability that any given viewer will report a post is extremely low — far lower than the probability of a like. A large negative weight ensures that when the model does predict a meaningful report probability, it can actually move the ranking. It is not a 1:1 cancellation against likes.

What is the difference between weights and adjustments?

Weights multiply predicted action probabilities to produce the base score. Adjustments (OonWeightFactor, AuthorDiversityDecay, ColdStart*) are applied after scoring to reshape the final ranking — discounts for out-of-network posts, decay for repeated authors, and boosts for new accounts.

How fresh is this data?

When our daily tracker detects a change in the repo, the live JSON is updated and this page reflects it within hours. If the tracker has not run or the live file is unavailable, the page falls back to a verified snapshot (marked "snapshot"). The source badge tells you which.

X Algorithm Weights — Live Reference | Qubax Growth Lab · Qubax AI