TogoLM
An open infrastructure layer for Togolese public data: a scraped corpus of laws, government records and press, a retrieval-augmented generation engine, and a REST API with JS/TS and Python client SDKs.
TogoLM builds an open data and AI infrastructure layer for Togo. A scraping pipeline collects more than 62,000 structured documents, including laws, government data, press coverage and education material, from over 55 sources, and indexes them for retrieval-augmented generation over the Togolese corpus.
The project exposes a public REST API that any developer or application can query, plus an admin API for corpus management, API key handling and query analytics. It also includes official JavaScript/TypeScript and Python SDKs, and is developing a Mistral 7B model fine-tuned to the Togolese context.
It runs on Python with FastAPI, PostgreSQL with the pgvector extension for embeddings, and Alembic for migrations, and targets developers, startups and institutions working with francophone West African data.