In development: 30% pre-trained data
Logo de PanamaLLMPanamaLLM

The first local AI model to empower Panama.

PanamaLLM is the first large language model trained with Panamanian data, designed to run locally and power AI solutions in the region.

We haven't closed it yet. If you're interested in trying the beta pilot, contact us:
Yalvi form
Available for macOS, Linux and Windows

Pre‑Training Data — Panama

We are getting started — overall progress: 30%

The map below shows the progress of data collection by province. Click a region to see detailed information.

Pre‑Training Data — Panama
Provinces in database
10
Total documents
321,000
Average completeness
30.0%
Top province
Panamá
Top 5 provinces by collection
Panamá
95,000
Chiriquí
45,000
Colón
35,000
Coclé
30,000
Veraguas
28,000
Quality and preparation
Duplicates removed12%
Spanish coverage92%
Average length (tokens)780
Preliminary figures subject to revision during pre-processing.
Governance and privacy
  • Personal data (PII) removal and human review by sample.
  • Compatible licenses and attribution registry by source.
  • Traceability and audit: each document retains origin and date.
  • Minimum access and encryption in transit and at rest.
Last validation:
Select a province on the map to view details.