datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
ClinFusion-Eval-Data
🏥 ClinFusion-Eval-Data
The Holistic Evaluation Suite for Vision-Centric Medical Multimodal LLMs
ClinFusion-Eval-Data is the unified evaluation corpus used to benchmark the ClinFusion model series (ClinFusion-8B, ClinFusion-32B). It packages 211,810 evaluation records spanning 22 public medical benchmarks into a single, consistently-formatted suite, together with 509 GiB of the underlying 2D images and native 3D CT volumes they refer to.
The goal is reproducibility:… See the full description on the dataset page: https://huggingface.co/datasets/Alibaba-DAMO-Academy/ClinFusion-Eval-Data.mfaqWe present the first multilingual FAQ dataset publicly available. We collected around 6M FAQ pairs from the web, in 21 different languages.Asclepius-Synthetic-Clinical-Notes
Asclepius: Synthetic Clincal Notes & Instruction Dataset
Dataset Summary
This dataset is official dataset for Asclepius (arxiv)
This dataset is composed with Clinical Note - Question - Answer format to build a clinical LLMs.
We first synthesized synthetic notes from PMC-Patients case reports with GPT-3.5
Then, we generate instruction-answer pairs for 157k synthetic discharge summaries
Supported Tasks
This dataset covers below 8 tasks
Named Entity… See the full description on the dataset page: https://huggingface.co/datasets/starmpcc/Asclepius-Synthetic-Clinical-Notes.mqaMQA is a multilingual corpus of questions and answers parsed from the Common Crawl. Questions are divided between Frequently Asked Questions (FAQ) pages and Community Question Answering (CQA) pages.clinical-trials
Clinical Trials Dataset
A comprehensive dataset of clinical trials sourced from ClinicalTrials.gov, featuring structured metadata, detailed study information, and pre-computed semantic embeddings for machine learning applications in biomedical research.
Dataset Description
This dataset provides a rich collection of clinical trial information systematically collected from the official ClinicalTrials.gov database. Each record contains detailed study metadata, eligibility… See the full description on the dataset page: https://huggingface.co/datasets/louisbrulenaudet/clinical-trials.medical-clinical-reasoning-sft-100k
Medical Clinical Reasoning SFT 100K
A synthetic supervised fine-tuning dataset of 100,000 high-quality medical and clinical reasoning conversations designed to train AI assistants capable of supporting clinical decision-making, documentation, and medical education.
Dataset Description
This dataset covers a broad spectrum of clinical practice scenarios across 10 medical specialty categories. Each record follows the ShareGPT conversation format with a detailed human… See the full description on the dataset page: https://huggingface.co/datasets/stindardlogic/medical-clinical-reasoning-sft-100k.protocolos-clinicos-br
Protocolos Clínicos BR
Paper | Code | Blog post
Brazilian Ministry of Health official clinical guidelines (PCDTs and related) plus the synthetic training corpus derived from them, used to adapt LLMs to Brazilian clinical knowledge. This dataset was introduced in the paper "Teaching LLMs Brazilian Healthcare: Injecting Knowledge from Official Clinical Guidelines".
Configurations
default — Original guidelines (raw text)
The 178 official Brazilian… See the full description on the dataset page: https://huggingface.co/datasets/hugo/protocolos-clinicos-br.ClimaQA
ClimaQA: An Automated Evaluation Framework for Climate Question Answering Models (ICLR 2025)
Check the paper's webpage and GitHub for more info!
The ClimaQA benchmark is designed to evaluate Large Language Models (LLMs) on climate science question-answering tasks by ensuring scientific rigor and complexity. It is built from graduate-level climate science textbooks, which provide a reliable foundation for generating questions with precise terminology and complex scientific theories.… See the full description on the dataset page: https://huggingface.co/datasets/Rose-STL-Lab/ClimaQA.ClimateFund
ClimateFund: An Annotated Dataset of Climate Mitigation Projects for Supporting Question Answering
This repository contains a dataset based on funding proposals of 21 climate mitigation projects, submitted to the Green Climate Fund (GCF).
Climate mitigation documentation is challenging to parse and understand, due to the length of this documents, their multi-modality (commonly comprising tables, figures and
free text), and their highly technical and domain-specific content.… See the full description on the dataset page: https://huggingface.co/datasets/JavierSanzCruza/ClimateFund.clinicalbr
ClinicalBr
ClinicalBr is the first bilingual (Portuguese–English) clinical-decision benchmark
built from 2,892 real Brazilian case reports drawn from 36 open-access
medical journals spanning 18 specialties. Every case is provided as a parallel
PT/EN pair and supports four evaluation tasks.
Please refer to the paper for full details on the tasks, methodology, and limitations.
Tasks & metrics
Task
Config
n / lang
Metric
Diagnosis retrieval
diagnosis
2,135… See the full description on the dataset page: https://huggingface.co/datasets/Giordanopsouza/clinicalbr.cli-commands-explained
Overview
This dataset is a collection of 16,098 command line instructions sourced from Commandlinefu and Cheatsheets. It includes an array of commands, each with an id, title, description, date, url to source, author, votes, and flag indicating if the description is AI generated. The descriptions are primarily authored by the original contributors, for entries where descriptions were absent, they have been generated using NeuralBeagle14-7B. Out of the total entries, 10,039… See the full description on the dataset page: https://huggingface.co/datasets/b-mc2/cli-commands-explained.clinical-trials-qa
Clinical Trials QA Dataset
A multi-tier question-answering benchmark for evaluating Retrieval-Augmented Generation (RAG) systems on clinical trial protocols from ClinicalTrials.gov.
Dataset Summary
This dataset provides question-answer pairs across four difficulty tiers, designed to benchmark RAG systems on real-world clinical trial documentation. Questions span four reasoning categories and require retrieval from protocol PDFs.
Key Features:
4 difficulty tiers based on… See the full description on the dataset page: https://huggingface.co/datasets/Parexel/clinical-trials-qa.PCFBench
PCFBench
Paper: arXiv:2608.27716 · Code: watershed-climate/pcfbench
Process-based Product Carbon Footprint benchmark for evaluating LLMs and
agents on the operational steps of life-cycle assessment (LCA): bill-of-
materials decomposition, mapping triage, ecoinvent process matching,
literature extraction of physical input rates, and total kgCO₂e
prediction against expert-grounded EPDs.
Tasks
ID
Task
Items
GT claims
Headline metric
1
Product decomposition… See the full description on the dataset page: https://huggingface.co/datasets/Watershed-Climate/PCFBench.qwen3.5-27b-cli-reasoning-3632x
Qwen3.5-27B CLI Reasoning 3632x
A synthetic reasoning dataset for CLI/terminal command assistance, distilled from Qwen3.5-27B with thinking mode enabled.
Each sample contains a realistic user scenario describing a terminal task, paired with the model's reasoning chain (<think>) and a structured JSON answer (command + description).
Dataset Summary
Source model
Qwen3.5-27B (DashScope API)
Samples
3,632
Thinking mode
Enabled (budget: 4096 tokens)… See the full description on the dataset page: https://huggingface.co/datasets/LocoreMind/qwen3.5-27b-cli-reasoning-3632x.epikg-clinicalbench
ClinicalBench: Assertion-Aware Clinical QA Benchmark
ClinicalBench is a benchmark for evaluating clinical question-answering systems on epistemic assertion reasoning over longitudinal patient records from MIMIC-IV.
It accompanies the paper:
ClinicalBench: Stress-Testing Assertion-Aware Retrieval for Longitudinal Clinical QA
(Paper under review; preprint forthcoming)
Benchmark Overview
400 questions across 9 assertion categories
43 MIMIC-IV patients with… See the full description on the dataset page: https://huggingface.co/datasets/alexstinard/epikg-clinicalbench.pentabrid-reproducibility
Pentabrid 27B: reproducibility package
Everything required to recompute the results of a controlled evaluation of fine-tuning
configurations for medical question answering. Openly available with no access
restrictions.
Contents
Path
Description
per_item/medxpertqa_*.jsonl
Per-item predictions for all six checkpoints on 2,450 MedXpertQA-Text items. Fields: id, gold, extracted_answer, correct, explicit_marker_present, n_markers, response_chars… See the full description on the dataset page: https://huggingface.co/datasets/Clinical-Reasoning-Hub/pentabrid-reproducibility.ClinVar-STXBP1-NLP-Datasetlanguage:
en
stxbp1_clinvar_curated
_ Curated STXBP1 and related variant records from ClinVar (24Million), ready for LLM and biomedical NLP applications._
Updated Jun 10th 2025. - Fields containing {null} or {} were removed.
Dataset Overview
A curated, LLM-friendly dataset of STXBP1 and related variant records from ClinVar, converted from ClinVar VCF and annotated for clinical, research, rare disease, and advanced AI applications.This resource is suitable for… See the full description on the dataset page: https://huggingface.co/datasets/SkyWhal3/ClinVar-STXBP1-NLP-Dataset.synthetic-clinical-notes-embedded
Synthetic Clinical Notes
This dataset is post-processed version of starmpcc/Asclepius-Synthetic-Clinical-Notes:
Turn into Alpaca format (instruction, input, and output)
Add embeddings for input and output columns using BAAI/bge-small-en-v1.5
Details
Sample Count
158k
Token Count
648m
Origin
https://figshare.com/authors/Zhengyun_Zhao/16480335
Source of raw data
PubMed Central (PMC) and MIMIC 3
Processing details
original, paper
Embedding Model… See the full description on the dataset page: https://huggingface.co/datasets/Technoculture/synthetic-clinical-notes-embedded.clinical-trial-outcomes-predictions
Clinical Trial Outcomes Prediction Dataset
A dataset of 1,366 binary forecasting questions about clinical trial outcomes, automatically generated and labeled using Lightning Rod Labs' Future-as-Label methodology.
Dataset Description
This dataset contains questions about pharmaceutical clinical trials from 2023-2024, paired with verified outcomes (success/failure). Each question asks whether a specific trial will meet its endpoints, receive FDA approval, or complete by a… See the full description on the dataset page: https://huggingface.co/datasets/3rdSon/clinical-trial-outcomes-predictions.clinical-persian-qa-ii
Clinical Question Answering Dataset II (Farsi)
This dataset contains more than 211k questions and more than 700k answers, all produced in written form. The questions were posed by ordinary Persian speakers (Iranians), and the responses were provided by doctors from various specialties.
Dataset Description
Question records without corresponding answers have been excluded from the dataset.
This dataset is NOT a part of Clinical Question Answering I dataset and is a whole… See the full description on the dataset page: https://huggingface.co/datasets/PerSets/clinical-persian-qa-ii.20Q20Qnigeria-clinical-guidelines-dataset
Dataset Card: NSTG 2022 Structured Clinical Dataset
Dataset Summary
A machine-readable, schema-unified JSON dataset of 270 clinical conditions derived from the Nigeria Standard Treatment Guidelines (NSTG) 2022. Each condition is represented as a structured JSON object covering clinical presentation, investigations, treatment protocols, differential diagnoses, complications, and prevention measures.
Designed for use in clinical AI/ML pipelines, decision support… See the full description on the dataset page: https://huggingface.co/datasets/chisomrutherford/nigeria-clinical-guidelines-dataset.psiddx-clinical-ddx
PsiDDx — Clinical Differential-Diagnosis Dataset (v5.2)
A decontaminated, calibrated differential-diagnosis training corpus for the Adaption Labs
AutoScientist Challenge (Healthcare). Each record pairs a patient presentation with a
ranked top-5 differential — common explanations first, each with an ICD-10 code, a calibrated
confidence, and the discriminating feature; rare diagnoses are kept as low-confidence
must-not-miss entries rather than over-called.
5,482 rows. Used to… See the full description on the dataset page: https://huggingface.co/datasets/shariqazeem/psiddx-clinical-ddx.clinical-rag-safety-gateway-20260904-dataset
Clinical RAG Safety Gateway Synthetic Dataset
Summary
This dataset contains 14 training examples and 4
held-out examples for Clinical assistants need retrieval, source attribution, and explicit abstention before answers reach care teams.
Every record is synthetic and includes:
input: query, event, or feature description
label: expected class, route, relation, or evidence category
context: synthetic supporting context
source: fictional source identifier
variant:… See the full description on the dataset page: https://huggingface.co/datasets/RKB109/clinical-rag-safety-gateway-20260904-dataset.ClinCheckBench
ClinCheckBench
A 4,101-sample benchmark for measuring adoption-blocking failure modes in clinical AI.
Dataset Description
ClinCheckBench spans seven clinical failure modes across three clinical workflow stages, evaluated on nine frontier LLMs with a three-tier scoring framework (deterministic, hybrid, LLM judge). The benchmark demonstrates that scoring methodology variance (40-80pp on factuality) can exceed between-model variance, and that every model exhibits a jagged… See the full description on the dataset page: https://huggingface.co/datasets/anonymous-clinbench/ClinCheckBench.clinicguide-faq-corpus
ClinicGuide FAQ corpus
Synthetic, generic clinic-style FAQs for a medical-tourism pre-consultation assistant. The corpus is not copied from a named hospital and is not a substitute for a real clinic's published policies.
Languages: English, Arabic (MSA), Persian.
Intended use
Retrieval-augmented answers for visa, stay, companion, hotel, airport transfer, starting-from costs, documents, booking, and “what to ask the doctor”
Safety / refusal examples: diagnosis… See the full description on the dataset page: https://huggingface.co/datasets/alirezaaminzadeh/clinicguide-faq-corpus.pocketgull-nih-who-clinical-dpo
📚 PocketGull NIH & WHO Clinical Preference DPO Dataset
Organization: PocketGull LLC (Oregon SOS: 258869891)Curator: Phillip Gear (CMS NPI: 1487569752 | ORCID: 0009-0008-1372-5381)License: Creative Commons Attribution 4.0 International (CC-BY-4.0)Open Science DOI: 10.5281/zenodo.20647514
📌 Dataset Summary
Gold-standard Direct Preference Optimization (DPO) chosen vs rejected pairs grounded in NIH MedQuAD, WHO mhGAP guidelines, and ClinicalTrials.gov protocols… See the full description on the dataset page: https://huggingface.co/datasets/philgear/pocketgull-nih-who-clinical-dpo.clinical-rag-safety-gateway-20260914-dataset
Clinical RAG Safety Gateway Synthetic Dataset
Summary
This dataset contains 14 training examples and 4
held-out examples for Clinical assistants need retrieval, source attribution, and explicit abstention before answers reach care teams.
Every record is synthetic and includes:
input: query, event, or feature description
label: expected class, route, relation, or evidence category
context: synthetic supporting context
source: fictional source identifier
variant:… See the full description on the dataset page: https://huggingface.co/datasets/RKB109/clinical-rag-safety-gateway-20260914-dataset.clinical-persian-qa-i
Clinical Question Answering Dataset I (Farsi)
This dataset contains approximately 50k questions and around 60k answers, all produced in written form. The questions were posed by ordinary Persian speakers (Iranians), and the responses were provided by doctors from various specialties.
Dataset Description
Question records without corresponding answers have been excluded from the dataset.
This dataset is NOT a part of Clinical Question Answering II dataset and is a complete… See the full description on the dataset page: https://huggingface.co/datasets/PerSets/clinical-persian-qa-i.ClinDiagnosES
Introducción
Este corpus se ha construido con ayuda de profesionales del sector de la salud de diversos ámbitos: cardiología, traumatología, urgencias, psiquiatría, neurología, dermatología, otorrino larongología, anestesia.
Guía de uso
El template para este dataset, con el fin de poder evaluar adecuadamente el rendimiento de LLMs sobre esta tarea, es el siguiente:
prompt_template="""A partir del caso clínico que se expone a continuación, tu tarea es la siguiente.
Como… See the full description on the dataset page: https://huggingface.co/datasets/LenguajeNaturalAI/ClinDiagnosES.