CheXpert CXR

CheXpert is a large dataset of 224,316 chest radiographs from 65,240 patients for automated chest X-ray interpretation. The dataset includes uncertainty labels for 14 medical observations extracted from radiology reports. It serves as a benchmark for developing and evaluating automated chest radiograph interpretation models.

MedGemma 4B IT from Google currently leads the CheXpert CXR leaderboard with a score of 0.481 across 1 evaluated AI models.

Paper
About this benchmark

What CheXpert CXR measures

CheXpert CXR is a image benchmark that evaluates large language models on healthcare and vision tasks. LLM Stats tracks 1 model on this benchmark, with a maximum possible score of 1. Current average across reported models is 0.5, with the leader reaching 0.5.

Compare leaders on the best AI for healthcare and best AI for vision leaderboards.

Publication

Paper
CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparison
Authors
Jeremy Irvin, Pranav Rajpurkar, Michael Ko, Yifan Yu, and 16 others
Published

Abstract

Large, labeled datasets have driven deep learning methods to achieve expert-level performance on a variety of medical imaging tasks. We present CheXpert, a large dataset that contains 224,316 chest radiographs of 65,240 patients. We design a labeler to automatically detect the presence of 14 observations in radiology reports, capturing uncertainties inherent in radiograph interpretation. We investigate different approaches to using the uncertainty labels for training convolutional neural networks that output the probability of these observations given the available frontal and lateral radiographs. On a validation set of 200 chest radiographic studies which were manually annotated by 3 board-certified radiologists, we find that different uncertainty approaches are useful for different pathologies. We then evaluate our best model on a test set composed of 500 chest radiographic studies annotated by a consensus of 5 board-certified radiologists, and compare the performance of our model to that of 3 additional radiologists in the detection of 5 selected pathologies. On Cardiomegaly, Edema, and Pleural Effusion, the model ROC and PR curves lie above all 3 radiologist operating points. We release the dataset to the public as a standard benchmark to evaluate performance of chest radiograph interpretation models. The dataset is freely available at https://stanfordmlgroup.github.io/competitions/chexpert .

GoogleMedGemma 4B IT leads with 48.1%.

Progress Over Time

Interactive timeline showing model performance evolution on CheXpert CXR

State-of-the-art frontier
Open
Proprietary

CheXpert CXR Leaderboard

1 models
ContextCostLicense
14B
Notice missing or incorrect data?

FAQ

Common questions about CheXpert CXR.

What is the CheXpert CXR benchmark?

CheXpert is a large dataset of 224,316 chest radiographs from 65,240 patients for automated chest X-ray interpretation. The dataset includes uncertainty labels for 14 medical observations extracted from radiology reports. It serves as a benchmark for developing and evaluating automated chest radiograph interpretation models.

What is the CheXpert CXR leaderboard?

The CheXpert CXR leaderboard ranks 1 AI models based on their performance on this benchmark. Currently, MedGemma 4B IT by Google leads with a score of 0.481. The average score across all models is 0.481.

What is the highest CheXpert CXR score?

The highest CheXpert CXR score is 0.481, achieved by MedGemma 4B IT from Google.

How many models are evaluated on CheXpert CXR?

1 models have been evaluated on the CheXpert CXR benchmark, with 0 verified results and 1 self-reported results.

Where can I find the CheXpert CXR paper?

The CheXpert CXR paper is available at https://arxiv.org/abs/1901.07031. The paper details the methodology, dataset construction, and evaluation criteria.

What categories does CheXpert CXR cover?

CheXpert CXR is categorized under healthcare and vision. The benchmark evaluates image models.

What is the best open-source model on CheXpert CXR?

MedGemma 4B IT by Google is the top-ranked open-source model on CheXpert CXR, with a score of 0.481 (rank #1).

How recent are the CheXpert CXR leaderboard results?

The CheXpert CXR leaderboard was last updated in June 2026 and currently includes 1 evaluated models.

More evaluations to explore

Related benchmarks in the same category

View all healthcare
MMLU-Pro

A more robust and challenging multi-task language understanding benchmark that extends MMLU by expanding multiple-choice options from 4 to 10, eliminating trivial questions, and focusing on reasoning-intensive tasks. Features over 12,000 curated questions across 14 domains and causes a 16-33% accuracy drop compared to original MMLU.

healthcare
127 models
MMLU

Massive Multitask Language Understanding benchmark testing knowledge across 57 diverse subjects including STEM, humanities, social sciences, and professional domains

healthcare
100 models
Humanity's Last Exam

Humanity's Last Exam (HLE) is a multi-modal academic benchmark with 2,500 questions across mathematics, humanities, and natural sciences, designed to test LLM capabilities at the frontier of human knowledge with unambiguous, verifiable solutions

visionmultimodal
82 models
MMMU

MMMU (Massive Multi-discipline Multimodal Understanding) is a benchmark designed to evaluate multimodal models on college-level subject knowledge and deliberate reasoning. Contains 11.5K meticulously collected multimodal questions from college exams, quizzes, and textbooks, covering six core disciplines: Art & Design, Business, Science, Health & Medicine, Humanities & Social Science, and Tech & Engineering across 30 subjects and 183 subfields.

healthcaremultimodal
62 models
MMMU-Pro

A more robust multi-discipline multimodal understanding benchmark that enhances MMMU through a three-step process: filtering text-only answerable questions, augmenting candidate options, and introducing vision-only input settings. Achieves significantly lower model performance (16.8-26.9%) compared to original MMMU, providing more rigorous evaluation that closely mimics real-world scenarios.

visionmultimodal
56 models
CharXiv-R

CharXiv-R is the reasoning component of the CharXiv benchmark, focusing on complex reasoning questions that require synthesizing information across visual chart elements. It evaluates multimodal large language models on their ability to understand and reason about scientific charts from arXiv papers through various reasoning tasks.

visionmultimodal
38 models