▌system-one

~ / open-implementations / trained-replicas / vegaml

vegaml

Typed decisions from a frozen LM plus a small physics engine, from Laya's author: calibrated probabilities, conformal answer sets, and an explicit abstain flag. 800M or 4B, 73K context, image inputs.

section
Open Implementations
group
Trained replicas
source
github.com
repo
nandhakishorm/vegaml
stars
196 ▕██████░░░░░░▏
7-day gain
+196
forks
25
language
Python
license
Apache-2.0
last pushed
oct 11
added to list
oct 12

## history

2026-10-040 → 196 stars2026-10-13

also in Trained replicas