RLTasksby ulam.ai
Models / OpenAI

OpenAI / MODEL EVIDENCE

GPT-3.5-1106

1 benchmarks · 8 observations · 1 sources

Scores at a glance

1 benchmarks · large dot: GPT-3.5-1106; small dots: closest models where reported

CRAFT-MD ↗37.5%
16.960.1

GPT-3.5-1106 · CRAFT-MD · Free response · Vignette accuracy · combined dataset ↗

7 more reported results

16.9% · GPT-3.5-1106 · CRAFT-MD · Free response · Multi-turn conversation accuracy · combined dataset · An evaluation framework for clinical use of large language models in patient interaction tasks — official CRAFT-MD leaderboard ↗

12.3% · GPT-3.5-1106 · CRAFT-MD · Free response · Single-turn conversation accuracy · combined dataset · An evaluation framework for clinical use of large language models in patient interaction tasks — official CRAFT-MD leaderboard ↗

17.4% · GPT-3.5-1106 · CRAFT-MD · Free response · Summarized conversation accuracy · combined dataset · An evaluation framework for clinical use of large language models in patient interaction tasks — official CRAFT-MD leaderboard ↗

65.9% · GPT-3.5-1106 · CRAFT-MD · Four-choice MCQ · Vignette accuracy · combined dataset · An evaluation framework for clinical use of large language models in patient interaction tasks — official CRAFT-MD leaderboard ↗

46.7% · GPT-3.5-1106 · CRAFT-MD · Four-choice MCQ · Multi-turn conversation accuracy · combined dataset · An evaluation framework for clinical use of large language models in patient interaction tasks — official CRAFT-MD leaderboard ↗

43.5% · GPT-3.5-1106 · CRAFT-MD · Four-choice MCQ · Single-turn conversation accuracy · combined dataset · An evaluation framework for clinical use of large language models in patient interaction tasks — official CRAFT-MD leaderboard ↗

50.9% · GPT-3.5-1106 · CRAFT-MD · Four-choice MCQ · Summarized conversation accuracy · combined dataset · An evaluation framework for clinical use of large language models in patient interaction tasks — official CRAFT-MD leaderboard ↗

Closest models by score

Similar positions on shared benchmark leaderboards. Smaller gaps are closer.

9.1 percentile-point gap

1 of 1 shared benchmark comparisons

18.2 percentile-point gap

1 of 1 shared benchmark comparisons

How similarity is measured

Mean absolute gap in percentile positions on shared benchmark settings; lower is closer. At least 60% overlap is required, with a minimum of three when available. Independent evaluations are used when there are at least three; otherwise published report comparisons are used. Repeated model observations within one setting are averaged before ranking. Writing-style traits and composite indices are excluded. This describes similarity, not an overall quality ranking.

Filter comparisons to this model →

Sources

Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.

  1. An evaluation framework for clinical use of large language models in patient interaction tasks — official CRAFT-MD leaderboard ↗

    Shreya Johri, Jaehwan Jeong, Benjamin A. Tran, Daniel I. Schlessinger, Shannon Wongvibulsin, Leandra A. Barnes et al. · 2025

    All 12 authors

    Shreya Johri, Jaehwan Jeong, Benjamin A. Tran, Daniel I. Schlessinger, Shannon Wongvibulsin, Leandra A. Barnes, Hong-Yu Zhou, Zhuo Ran Cai, Eliezer M. Van Allen, David Kim, Roxana Daneshjou, Pranav Rajpurkar

Names used in source tables

GPT-3.5-1106