Now, in my "C:\Users\**\Axiom\Contribute i have a file .bin, seed_data of 10mb.
I have also some bin files in the .gradients folder.

Yesterday i thinked to put some pdf from CERN into the Contribute folder, but there is these files...

This is from the news section on the main website:

Feb 8, 2026
v3.51: Seed training data - workers now automatically download 10MB of diverse public domain text (Project Gutenberg) on startup, ensuring all volunteers have quality training data from day one. Data refreshes every 24 hours with a unique random mix from a 10GB+ server pool. Gossip gradient improvements: per-host naming, 10-minute decay, random peer gradient applied at task start. Entropy filter tightened to 7.0 bits/byte. Model evaluation now updates every 5 minutes.

So the way I understand it is that we don't have to provide our own data anymore?

Never had to. I've only seen the message when starting the project on a client.

Without data the tasks cannot get trained so you probably always got 0 points for your tasks?

I have credit for clients I've never added any files to the contribute folder.

The same host can have some tasks at 0, the next with credit. The next 0. While yesterday the host had every task with credit.

It seems like 3.51 introduced an automatic sample DL so nothing is needed from the user.

From the About Axiom Distributed AI page:

The 85/15 Split: 85% of learning happens locally on your machine, while 15% comes from gossip synchronization with peers training other experts.

If you contribute files, axiom will spend 15% of it's time learning from peer files anyway. So I suppose if you don't contribute files, axiom will spend 100% of it's time on peer files. Your own files aren't mandatory.

---