datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
SWE-smith-javamigration-bench-java-full
MigrationBench
1. 📖 Overview
🤗 MigrationBench
is a large-scale code migration benchmark dataset at the repository level,
across multiple programming languages.
Current and initial release includes java 8 repositories with the maven build system… See the full description on the dataset page: https://huggingface.co/datasets/AmazonScience/migration-bench-java-full.Java-Code-Large-text-onlyJava-Code-Large
Java-Code-Large is a large-scale corpus of publicly available Java source code comprising more than 15 million java codes. The dataset is designed to support research in large language model (LLM) pretraining, code intelligence, software engineering automation, and program analysis.
By providing a high-volume, language-specific corpus, Java-Code-Large enables systematic experimentation in Java-focused model training, domain adaptation, and downstream code understanding tasks.… See the full description on the dataset page: https://huggingface.co/datasets/XEUIPR/Java-Code-Large-text-only.n-hypo-javaJavaScript-Code-LargeJavaScript-Code-Large
JavaScript-Code-Large is a large-scale corpus of JavaScript source code comprising around 5 million JavaScript files. The dataset is designed to support research in large language model (LLM) pretraining, code intelligence, software engineering automation, and program analysis for the JavaScript ecosystem.
By providing a high-volume, language-specific corpus, JavaScript-Code-Large enables systematic experimentation in JavaScript-focused model training, domain adaptation… See the full description on the dataset page: https://huggingface.co/datasets/ajibawa-2023/JavaScript-Code-Large.javascript-github-codecornstack-java-v1
CoRNStack Python Dataset
The CoRNStack Dataset, accepted to ICLR 2025, is a large-scale high quality training dataset specifically for code retrieval across multiple
programming languages. This dataset comprises of <query, positive, negative> triplets used to train nomic-embed-code,
CodeRankEmbed, and CodeRankLLM.
CoRNStack Dataset Curation
Starting with the deduplicated Stackv2, we create text-code pairs from function docstrings and respective code. We filtered out… See the full description on the dataset page: https://huggingface.co/datasets/nomic-ai/cornstack-java-v1.java_PRscornstack_java_ru_enThe part of CoRNStack Dataset translated into Russian. Translation was done with Qwen3 model.
Samples that satisfy the dual consistency filtering condition (samples where the document_rank is 0 or 1 and document_score > 0.7) were translated.
Source code you can find here. For support: fedor.yaronskiy@gmail.com
Stackless_Java_V2
Dataset Summary
This is the dataset used for the training of the AP-MAE models, it is a subset of The Heap, we release it for reproducability.
AIXCC-Java-Challenge
🧠 AIXCC Challenge Benchmark – Java Challenges
📌 Overview
AIXCC Challenge Benchmark (Java Challenges) is a curated subset of the full AIXCC Challenge Benchmark focused solely on C-based challenges.
This benchmark is built upon the official AIXCC Challenge. Each Java challenge consists of either a delta (focused diff) or full (whole project) test case.
📎 Reference Implementation
This benchmark is designed to work with our open-source CRS system:… See the full description on the dataset page: https://huggingface.co/datasets/Kitxuuu/AIXCC-Java-Challenge.deewaiREALCN-training
Repo
git@hf.co:datasets/telcom/deewaiREALCN-training
DeewaiREALCN Training Data
Image–text pairs for training captioning or vision–language models. Each image is a 1024×1024 RGB JPEG portrait with a short English description.
Contents
data/train/: 9,000 pairs for training.
images/: JPEG files (090000.jpg, …).
captions.jsonl: one JSON object per line with file_name and text.
data/val/: 1,000 pairs for validation with the same layout.
Example… See the full description on the dataset page: https://huggingface.co/datasets/javadtaghia/deewaiREALCN-training.codeparrot-javamigration-bench-java-selected
MigrationBench
1. 📖 Overview
🤗 MigrationBench
is a large-scale code migration benchmark dataset at the repository level,
across multiple programming languages.
Current and initial release includes java 8 repositories with the maven build system… See the full description on the dataset page: https://huggingface.co/datasets/AmazonScience/migration-bench-java-selected.repobench_java_v1.1
RepoBench v1.1 (Java)
Introduction
This dataset presents the Java portion of RepoBench v1.1 (ICLR 2024). The data encompasses a collection from GitHub, spanning the period from October 6th to December 31st, 2023. With a commitment to data integrity, we've implemented a deduplication process based on file content against the Stack v2 dataset (coming soon), aiming to mitigate data leakage and memorization concerns.
Resources and Links
Paper
GitHub
Dataset… See the full description on the dataset page: https://huggingface.co/datasets/tianyang/repobench_java_v1.1.JavaScript-Code-LargeJavaScript-Code-Large
JavaScript-Code-Large is a large-scale corpus of JavaScript source code comprising around 5 million JavaScript files. The dataset is designed to support research in large language model (LLM) pretraining, code intelligence, software engineering automation, and program analysis for the JavaScript ecosystem.
By providing a high-volume, language-specific corpus, JavaScript-Code-Large enables systematic experimentation in JavaScript-focused model training, domain adaptation… See the full description on the dataset page: https://huggingface.co/datasets/thedruid831/JavaScript-Code-Large.JavaScript-Code-LargeJavaScript-Code-Large
JavaScript-Code-Large is a large-scale corpus of JavaScript source code comprising around 5 million JavaScript files. The dataset is designed to support research in large language model (LLM) pretraining, code intelligence, software engineering automation, and program analysis for the JavaScript ecosystem.
By providing a high-volume, language-specific corpus, JavaScript-Code-Large enables systematic experimentation in JavaScript-focused model training, domain adaptation… See the full description on the dataset page: https://huggingface.co/datasets/liuhangbiao/JavaScript-Code-Large.Methods2Test_java_unit_test_code
Dataset Description
Microsoft created this large dataset of Java Junit test cases with its corresponding focal methods.
It contains 780k pairs of JUnit test cases and focal methods which were extracted from a total of 91K
Java open source project hosted on GitHub.
The mapping between test case and focal methods are based heuristics rules and Java developer's best practice.
More information could be found here:
methods2test Github repo
Methods2Test: A dataset of focal methods… See the full description on the dataset page: https://huggingface.co/datasets/jitx/Methods2Test_java_unit_test_code.Java-Code-LargeJava-Code-Large
Java-Code-Large is a large-scale corpus of publicly available Java source code comprising more than 15 million java codes. The dataset is designed to support research in large language model (LLM) pretraining, code intelligence, software engineering automation, and program analysis.
By providing a high-volume, language-specific corpus, Java-Code-Large enables systematic experimentation in Java-focused model training, domain adaptation, and downstream code understanding tasks.… See the full description on the dataset page: https://huggingface.co/datasets/ajibawa-2023/Java-Code-Large.JavaScript-Code-LargeJavaScript-Code-Large
JavaScript-Code-Large is a large-scale corpus of JavaScript source code comprising around 5 million JavaScript files. The dataset is designed to support research in large language model (LLM) pretraining, code intelligence, software engineering automation, and program analysis for the JavaScript ecosystem.
By providing a high-volume, language-specific corpus, JavaScript-Code-Large enables systematic experimentation in JavaScript-focused model training, domain adaptation… See the full description on the dataset page: https://huggingface.co/datasets/Ujjwal-Tyagi/JavaScript-Code-Large.Stack2Graph_KG_java
Java StackOverflow Knowledge Graph
Summary
This Hugging Face dataset repository contains the Java shard of the Stack2Graph StackOverflow Knowledge Graph.
Hugging Face uses one dataset repository per programming language, so this repository is directly cloneable without an extra top-level archive wrapper.
The artifact is optimized for graph-based retrieval, SPARQL analytics, and retrieval-augmented question answering over Stack Overflow content.
Stack2Graph source:… See the full description on the dataset page: https://huggingface.co/datasets/Mo7art/Stack2Graph_KG_java.Java-codeSWE-smith-java-6704-filtered-for-problem-statementsjavascript-mediumcornstack-javascript-v1
CoRNStack Javascript Dataset
The CoRNStack Dataset, accepted to ICLR 2025, is a large-scale high quality training dataset specifically for code retrieval across multiple
programming languages. This dataset comprises of <query, positive, negative> triplets used to train nomic-embed-code,
CodeRankEmbed, and CodeRankLLM.
CoRNStack Dataset Curation
Starting with the deduplicated Stackv2, we create text-code pairs from function docstrings and respective code. We filtered… See the full description on the dataset page: https://huggingface.co/datasets/nomic-ai/cornstack-javascript-v1.Java-Code-LargeJava-Code-Large
Java-Code-Large is a large-scale corpus of publicly available Java source code comprising more than 15 million java codes. The dataset is designed to support research in large language model (LLM) pretraining, code intelligence, software engineering automation, and program analysis.
By providing a high-volume, language-specific corpus, Java-Code-Large enables systematic experimentation in Java-focused model training, domain adaptation, and downstream code understanding tasks.… See the full description on the dataset page: https://huggingface.co/datasets/Ujjwal-Tyagi/Java-Code-Large.stark-nano-java-corpusJavaTestGen
JavaTestGen
Task description
Java TestGen is a benchmark designed to evaluate code generation models' ability to generate Java unit tests. Tasks involve generating unit tests based on provided Java source code and repository context. Dataset contains 227 tasks.
Evaluated skills: Instruction Following, Code Perception, Completion, Testing
Contributors: Dmitry Salikhov, Pavel Zadorozhny, Pavel Adamenko, Rodion Levichev, Aidar Valeev, Dmitrii Babaev
Motivation… See the full description on the dataset page: https://huggingface.co/datasets/MERA-evaluation/JavaTestGen.Java-GitHub-Codesgithub-file-programs-dataset-java