FOR PHARMA
FOR AI LABS
Longitudinal clinical datasets, training environments and private evaluation.
RESEARCH & OPEN TOOLS
Small models. Unreasonably specific interests.
We sent a 4B model to the Gym. It developed a taste for healthcare admin.
CRS-4B is fine-tuned only on public synthetic Gym episodes and evaluated on the private split.
System One for speech. Ask typed questions about short audio windows and get structured decisions.
Eight-second windows. Term scores are uncalibrated.
Open the report. Find the encounter. Put the right thing in the right place. We’ll check the chart.
727 public tasks · 5 families · 315 Hard · 126 image · 189 private
Scored on the final record state. Offline. No LLM judge.
14 models on 99 private tasks. Three attempts per task. November 2026.
01
CSV, HL7 v2, broken-table PDFs, superseded reports, non-standard units, ambiguous dates and same-name patients.
02
Difficulty bands for every task.
03
Calculators implemented from the primary publications, including the correct refusal when inputs are missing.
04
A synthetic web EHR with screenshots, coordinates, an accessibility tree and human approval steps, scored on the final FHIR state.
05
Published as a versioned dataset.
FOR DATA PARTNERS
Source datasets for your AI lab customers, with agreed cohorts, quality criteria and delivery cadence.
Tell us what data you need. We’ll shape the cohort, format and delivery around it.
