peterBlue75

I can't comment on concrete sizes of future models, partly because it's not decided yet. But: Pre-training for Kolibri only started in August so there's a high chance continued post-training - which we plan to do - will yield some nice checkpoints. We believe this size and sparsity allows achieving great inference throughput at reasonable levels of intelligence.

Related thread on the tooling:

Model Training as Code - https://news.ycombinator.com/item?id=48673450 - June 2026 (24 comments)

Thanks! What matters to us right now is not so much our current position, but the velocity with which we’re moving. Kolibri 1 is the first measurement of our position, there’ll be more. We want to build this in Europe and this is only the start. The report also shows our focus on building out proper tooling with our Model Factory.

The weights have been published to HF on an apache2.0 license, and the tech report is quite extensive

The thing to note here, besides the transparency and the fact that it’s actually a good model that also works well on coding and agentic tasks, is that it’s the first release by a team formed less than a year ago, with a strong focus on iteration velocity. There’s more to come.

disclaimer: I‘m part of the training team, happy to answer any questions

The thing to note here, besides the transparency and the fact that it’s actually a good model that also works well on coding and agentic tasks, is that it’s the first release by a team formed less than a year ago, with a strong focus on iteration velocity. There’s more to come.

disclaimer: I‘m part of the training team, happy to answer any questions

can’t wait to tell my CISO that our production binary depends on Sam Altman’s current mood and a $20/month Plus subscription

Atmospheric waves are quite fascinating. Crazy to see a glider utilize this over such a distance:

https://www.weather.gov/source/zhu/ZHU_Training_Page/Miscell...