+
Research Product
Horizon: Research Without the Conversational Padding
Horizon turns a question into a structured research artifact: direct findings, visible uncertainty, and evidence you can inspect without the sentimental filler.
Product
Inside Horizon
Notes from the systems, models, and research workflows behind North ML. See how Horizon turns questions into direct, inspectable output.
Open Horizon+
Retrieval and synthesis designed to keep claims separable from conclusions.
Read Research note 0.2 ↗+
Compact model experiments focused on capability per parameter and visible API cost.
Read Systems note 0.2 ↗+
A workflow that prioritizes usable output over a long conversational performance.
Read Product note 0.2 ↗Useful output. Visible limits. Less conversational performance.
<1B parameter model experiments
1,000+ words in research threads
<$3 target for mid-sized synthesis