Skip to content
Lorraine logo

Clinical AI, calibrated for SA.

A small catalogue of purpose-built clinical models. Each card describes what the model is for, how it should be used, and where its limits are.

/architecture

Summary
serving
Lor-1 v1.0
app region
af-south-1 (Cape Town)
inference
GPU · United States
bench
LorBench · SA
strengths
dosing · ddx
api
openai-compatible
How Lor-1 is served today. Not a live status feed — service health is not published on this page.

/flagship

Lor-1

The clinical reasoning model for South Africa.

A clinical foundation model calibrated to reason inside South African clinical conventions — SA drug names, the tiered Essential Medicines List, national treatment guidelines, and local referral pathways. Served behind an OpenAI-compatible API.

  • Differential reasoning grounded in SA Standard Treatment Guidelines.
  • Drug lookup aware of EML tier and care-level availability.
  • Structured summarisation and extraction from clinical notes.
  • Explicit abstention when SA guidance is insufficient.
Open model card

/benchmark

Strongest lifts
DosingSA drug names, dosing schedules, and formulary conventions.
Geographic fit
SA-nativeLocal drug names, formulary tiers, and referral pathways by default.
Serving
One endpointOpenAI-compatible. Topology not published.
A specimen slide held on a microscope stage, washed in violet light

Calibration is measured on SA source material, one question at a time.

On the roadmap

What’s next.

Published so customers can plan around it. Nothing ships until it clears the same benchmark bar as Lor-1.

/research

Lor-1 Vision

Multimodal extension of Lor-1 for clinical imaging in SA settings — chest X-ray triage, paediatric growth charts, and diagnostic photograph review with SA-grounded interpretation.

/research

Lor-1 Small

A smaller, edge-friendly variant of Lor-1 for low-bandwidth and on-prem deployments. Same SA calibration, a fraction of the infrastructure footprint.

How we publish

Design commitments.

04 principles

  1. 01

    Clinical scope

    Each card states intended use, user assumptions, and the clinical boundaries where a qualified clinician must remain responsible for the decision.

  2. 02

    Measured, not asserted

    Lor-1 is promoted only on a measured lift on LorBench, our internal SA clinical benchmark. Its card sets out what LorBench tests and how the promotion gate works; the figures behind it are shared under enterprise review rather than published here.

  3. 03

    Honest limitations

    We publish known failure modes, recall ceilings, and category weaknesses alongside the headline numbers. Silence about limitations is not a marketing feature.

  4. 04

    Controlled disclosure

    Deployment-specific evidence packs, evaluation reports, and architecture notes are shared through enterprise review under NDA — not by default.

Public cards intentionally omit training recipes, dataset composition, serving topology, and internal codenames. Enterprise customers get a deeper technical appendix under NDA.

Request enterprise review