Observed arrival · 2026-09-26
Kvaxir: a local, no-code modeling studio
Kvaxir presents a Windows app for collecting structured data, training prediction models locally, and exporting them for independent use.
- For
- Windows users teaching or building models from device and structured data
- Worth noticing
- The page discloses that code signing is not yet implemented and provides SHA-256 checksums and installer-verification instructions.
Field notes
The listed Windows workflow takes structured data from CSV or serial, Bluetooth, TCP/UDP, and MQTT sources through labeling, local training, evaluation, and export. The page says exports are standard AutoGluon or pickle packages loadable independently with Python; it also separates offline modeling from optional AI features, which may send column names, row counts, and feature summaries unless pure-local mode is enabled. Version 0.3.11 is labeled public beta, and code signing is listed as not yet implemented.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
One card from the complete issue