"We have open-sourced our new algorithm, powered by the same transformer architecture as xAI's Grok model."
Oh I see it is not meant to be built really. Some code is omitted.
looks like this is the "for you" feed, once again shared without weights so we only have so much visibility into the actual influence of each trait.
"We have eliminated every single hand-engineered feature and most heuristics from the system. The Grok-based transformer does all the heavy lifting by understanding your engagement history (what you liked, replied to, shared, etc.) and using that to determine what content is relevant to you." aka it's a black box now.
the README is actually pretty nice, would recommend reading this. it doesnt look too different form Elon's original code review tweet/picture https://x.com/elonmusk/status/1593899029531803649?lang=en
sharing additonal notes while diving through the source: https://deepwiki.com/xai-org/x-algorithm
and a codemap of the signal generation pipeline: https://deepwiki.com/search/make-a-map-of-all-the-signals_3d...
- Phoenix (out of network) ranker seems to have all the interesting predictive ML work. it estimates P(favorite), P(reply), P(repost), P(quote), P(click), P(video_view), P(share), P(follow_author), P(not_interested), P(block_author), P(mute_author), P(report) independently and then the `WeightedScorer` combines them using configurable weights. there's an extra DiversityScore and OONScore to add some adjustments but again dont know the weights https://deepwiki.com/xai-org/x-algorithm/4.1-phoenix-candida... - other scores of interest: photo_expand_score, and dwell_score and dwell_time. share via copy, share, and share via dm are all obviously "super like" buttons.
- Two-Tower retrieval uses dot product similarity between user features/engagement (User Tower) and normalized embeddings for all items (Candidate Tower). but when you look into the code and considering that this is probably the most important model for recommendations quality.... it's maybe a little disappointing that its a 2 layer MLP? https://deepwiki.com/search/what-models-are-used-for-user_98...
- Grok-1 JAX transformer (https://github.com/xai-org/x-algorithm/blob/main/phoenix/REA...) uses special attention masking that prevents candidates from attending to each other during inference. Each candidate only attends to the user context (engagement history). This ensures a candidate's score is independent of which other candidates are in the batch, enabling score consistency and caching. nice image here https://github.com/xai-org/x-algorithm/blob/main/phoenix/REA...
- kind of nice usage of Rust traits to create a type safe data pipeline. look at this beautiful flow chart https://deepwiki.com/xai-org/x-algorithm/3-candidate-pipelin... and the "Field Ownership pattern" https://deepwiki.com/xai-org/x-algorithm/3.6-scorer-trait#fi...
- the ten pre-scoring filters are minorly interesting, nothing super surprising here apart from AgeFilter (https://deepwiki.com/xai-org/x-algorithm/4.6.1-agefilter) which I guess means beyond a certain max_age (1 day?) nothing ever shows up on For You. surprising to have a simple flat cutoff vs i guess the alternative of an exponential aging algorithm.
- videoduration hydrator explicitly prioritizes video duration (https://deepwiki.com/xai-org/x-algorithm/4.5.6-videoduration...) but we dont know in what direction... do you recommend shorter or longer videos? and why a hydrator for what is presumably a pretty static property?
open questions from me
1. how large is the production reranker? default param count is here https://deepwiki.com/search/how-many-params-is-the-transfo_c... but that gives no indication. the latency felt ultra high initially last year and seems to have come down some, what budget are we working with?
2. can we make the retrieval better? i dont have a tooon of confidence in the User Tower / Candidate Tower system - is this SOTA (it's probably not - see how youtube does codebook semantic id's https://www.youtube.com/watch?v=LxQsQ3vZDqo&list=PLcfpQ4tk2k... )
3. no a/b testing / rollout infrastructure?
4. so many hydration subsystems - is this brittle?
Someone explain.
They most likely have some secret sauce that they don't release to public.
X has content moderation that relies on a mix of AI and human review, focusing on automated systems and user reports. There’s less emphasis on account suspensions and more on reach restriction, alongside community-led moderation like "Community Notes"
Plus they had done this before and no real competitor raised since last time they did it. So why not do it again.
xAI likely needs both more than usual nowadays.
Honestly, this looks like a PoC - Proof of Concept. They've open sourced what used to be a PoC at one point.
They're eating the code. They're eating the algorithms.
Is Grok not an LLM? Or do they have other models under that brand?
> Is Grok not an LLM?
Transformer is the underlying technology for (most) LLMs (GPT stands for “Generative Pre-Trained Transformer”)
I'm sure there's many examples but here's the first Google search result: https://www.theguardian.com/us-news/2025/nov/12/elon-musk-gr...
Meanwhile the people making a fuss about it are the same people that voted against investigating the recent child abuse scandal in the UK.
I have character and I'm German who knows his history.
I sincerely hope that the main stream media does not fall for this and calls it out. It's not rocket science. It's really really simple - this is not good for anyone.
If I "Open Source" windows 11 but lie and put some other junk there then I can't CLAIM to have open sourced windows 11 now can I?
You can claim the open source code isn’t Windows 11, but you can’t complain the code isn’t open source.
I'm confused what this has to do with "open source" or how it affects public perception.
I agree with you that it's totally possible to lie about what is actually running in production and that sharing some code doesn't mean it's that code, but how is this a new problem?
So in the end are we going by the OSI's definition of Open Source, or not? Can we make up our mind please?
Every time anyone posts here even a slightly modified Open Source license (e.g. a MIT license with an extra restriction that prevents megacorporations from using it but doesn't affect anyone else) people come out of the woodwork with their pitchforks screaming "this is not Open Source!", and insist that the Open Source Definition decides what is Open Source or not, and not to call anything which doesn't meet that definition "Open Source".
And yet here we are with a repository licensed under an actually Open Source license, and suddenly this is the most upvoted comment, and now people don't actually care about the Open Source Definition after all?
Either we go by the OSI's definition, in which case this is open source, regardless of what you think the motivations are for opening up this code, or we go by the "vibes" of whether it feels open source, in which case a modified MIT license which prohibits companies with a trillion+ market cap from using it is also open source.
``` @route("/"): def main(): return "hello world" ```
What does that give us? We can't run this to host our own hackernews as it's clearly not runnable. We can't really learn anything from this as it doesn't not represent any real reality. Maybe it's a fun reading exercise but that's about it.
Open source means that I can take source and run it to ensure it's trusted. Ascii characters being visible on my screen is just a nice byproduct of this goal.
They pretend to be transparent about their algorithms while denying researchers access to their API through exorbitant pricing and severely limited quotas.
I am sceptical of Musk, but this seems to be a legitimate transparency move.