RxScribe Bench - Correctness (Clinical and Non-Clinical Fields): leaderboard

Metric: Field-level F score (%; harmonic mean of credit-weighted precision and recall per run, averaged over the three runs and then over images; diagnoses, symptoms, vitals, history, allergies and tests plus prescriber, facility and date fields, scored by ANLS string similarity; 200 handwritten Indian outpatient prescription images, three independent cold runs per image, image and JSON schema only, deterministic field-by-field scoring against human ground truth with no model judge). Source: arxiv.org. Saturation forecast: Around September 2027. 4 models tracked.

Top models

#ModelScore
1Gemini 3.1 Pro (Preview)78.94
2Muse Spark 1.278.87
3GPT-5.6 Sol77.84
4Claude Opus 575.59

Interactive version: theaggregate.ai/benchmark?slug=rxscribe-bench-correctness-clinical-and-non-clinical-fields · How It Works · Data refreshed daily, snapshot 2026-09-26.