User-owned local path
Your machine.
Your model.
The Local Companion binds to loopback, authenticates the browser with a pairing token, and routes compatible local text inference from any entitled box. GoldCap alone prepares approved training data and starts user-configured compatible training.

CURRENT RELEASE
Download decacap-local-companion.zipVersion 1.0.0
Released 2026-07-23 · 9.2 KB · Python 3.11+
SHA-256 60ae0416c913a725b54be8e33ce0fc2d2867c62150b86dbb431ae52b6827d3f9Install + pair
- Verify the SHA-256 above.
- Extract the ZIP to a user-owned folder.
- Install Python 3.11 or newer and a local runtime.
- Run
start-windows.bat. - Enter the printed loopback endpoint and token in Workbench Local Settings.
- Run health, model-list, and short inference tests.
Install + pair
- Verify the SHA-256 above.
- Extract the ZIP to a user-owned folder.
- Install Python 3.11 or newer and a compatible runtime.
- Run
python3 decacap_companion.py. - Pair from Workbench Local Settings.
- Run health, model-list, and short inference tests.
What this build contains
Removed fixed inference timeout
Per-user loopback pairing
Ollama and compatible local text-runtime support
Explicit local data preparation and training lifecycle controls
- Stop the current companion process.
- Back up
decacap-companion.jsonand the local data folder. - Verify and extract the new ZIP into a separate folder.
- Carry forward only reviewed local configuration.
- Restart, re-pair if the token changed, and rerun health/inference.
- Stop the companion process.
- Remove the extracted program files.
- Remove the device-local Workbench endpoint/token from Local Settings.
- Keep or delete local datasets, models, adapters, and logs according to your own retention decision.
- Uninstall the separate model runtime only if no other application uses it.