m1-7B-23K: benchmark results
Provider: Other.
Aggregate Rating 1420 ± 33, rank #5166 of 10916 rated models, from 13 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| CureMed-Bench | 42.9 | Logical accuracy (%; share of final answers a GPT-4o judge m | 64.6 |
| MR-Bench - Procedure Selection | 57.5 | Accuracy (%, times 100) at picking the admission's procedure | 47.8 |
| MR-Bench | 39.1 | Accuracy (%, times 100), mean of the medication-imputation a | 39.1 |
| CureMed-Bench - Language Consistency | 17.61 | Language consistency (%; share of final answers written in t | 27.1 |
| EHRBench | 46.08 | Overall Acc (%) (self-reported) | 20 |
| EHRBench - MIMIC-IV | 46.99 | Accuracy (%) on the questions built from MIMIC-IV records, E | 20 |
| EHRBench - PROMOTE | 45.14 | Accuracy (%) on the questions built from PROMOTE records, EH | 20 |
| EHRBench - Treatment | 63.07 | Accuracy (%) on the treatment-decision questions, EHR-ground | 20 |
| EHRBench - MIMIC-III | 46.5 | Accuracy (%) on the questions built from MIMIC-III records, | 16.7 |
| EHRBench - Diagnosis | 38.42 | Accuracy (%) on the diagnosis-decision questions (complete a | 13.3 |
| EHRBench - Prognosis | 36.74 | Accuracy (%) on the prognosis-decision questions (predict a | 13.3 |
| MR-Bench - Medication Imputation | 20.6 | Accuracy (%, times 100) at picking the masked medication amo | 8.7 |
Interactive version: theaggregate.ai/model?slug=m1-7b-23k · How It Works · Data refreshed daily, snapshot 2026-11-08.