Maternity Case Record Digitisation - Accuracy without Nulls: leaderboard

A 2026 benchmark of multimodal LLMs for digitising a handwritten South African maternity case record: each form image is extracted into a JSON template and compared field by field with reference JSON files. This board is field-level exact-match accuracy in percent over populated fields only, median over the real registry forms; higher is better.

Metric: Field-level accuracy over populated fields only (from 0-1, times 100; median over forms): share of filled-in fields whose normalised extraction exactly matches the ground truth, on the real redacted Maternity Case Records (49 forms for Gemini and OpenAI models; 46 for Claude) of the UBOMI BUHLE Pregnancy Exposure Registry (South Africa), one form image per call with the optimised extraction prompt and JSON template, temperature 0, two attempts at valid JSON; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 9 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 3.1 Pro (Preview)78#99
2GPT-5.476#135
3Gemini 2.5 Pro74#276
4Gemini 2.5 Flash73#463
5Gemini 3.1 Flash Lite73#371
6Claude Sonnet 4.664#173
7Claude Opus 4.657#113
8Claude Haiku 4.551#583
9GPT-5 Mini31#364

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=maternity-case-record-digitisation-accuracy-without-nulls · How It Works · Data refreshed daily, snapshot 2026-11-21.