nvidia/Nemotron-Pretraining-Code-v3

1---2license: cc-by-4.03task_categories:4- text-generation5tags:6- text7- pre-training8- human9- legal10- Nemotron_3_Ultra11language:12- code13size_categories:14- 100M<n<1B15configs:16- config_name: Nemotron-Code-Metadata17  data_files:18  - path:19    - Nemotron-Code-Metadata/part_00000.parquet20    - Nemotron-Code-Metadata/part_00001.parquet21    - Nemotron-Code-Metadata/part_00002.parquet22    - Nemotron-Code-Metadata/part_00003.parquet23    - Nemotron-Code-Metadata/part_00004.parquet24    - Nemotron-Code-Metadata/part_00005.parquet25    - Nemotron-Code-Metadata/part_00006.parquet26    - Nemotron-Code-Metadata/part_00007.parquet27    - Nemotron-Code-Metadata/part_00008.parquet28    - Nemotron-Code-Metadata/part_00009.parquet29    - Nemotron-Code-Metadata/part_00010.parquet30    - Nemotron-Code-Metadata/part_00011.parquet31    - Nemotron-Code-Metadata/part_00012.parquet32    - Nemotron-Code-Metadata/part_00013.parquet33    - Nemotron-Code-Metadata/part_00014.parquet34    - Nemotron-Code-Metadata/part_00015.parquet35    - Nemotron-Code-Metadata/part_00016.parquet36    - Nemotron-Code-Metadata/part_00017.parquet37    - Nemotron-Code-Metadata/part_00018.parquet38    - Nemotron-Code-Metadata/part_00019.parquet39    - Nemotron-Code-Metadata/part_00020.parquet40    - Nemotron-Code-Metadata/part_00021.parquet41    - Nemotron-Code-Metadata/part_00022.parquet42    - Nemotron-Code-Metadata/part_00023.parquet43    - Nemotron-Code-Metadata/part_00024.parquet44    - Nemotron-Code-Metadata/part_00025.parquet45    - Nemotron-Code-Metadata/part_00026.parquet46    - Nemotron-Code-Metadata/part_00027.parquet47    - Nemotron-Code-Metadata/part_00028.parquet48    - Nemotron-Code-Metadata/part_00029.parquet49    - Nemotron-Code-Metadata/part_00030.parquet50    - Nemotron-Code-Metadata/part_00031.parquet51    - Nemotron-Code-Metadata/part_00032.parquet52    - Nemotron-Code-Metadata/part_00033.parquet53    - Nemotron-Code-Metadata/part_00034.parquet54    - Nemotron-Code-Metadata/part_00035.parquet55    - Nemotron-Code-Metadata/part_00036.parquet56    - Nemotron-Code-Metadata/part_00037.parquet57    - Nemotron-Code-Metadata/part_00038.parquet58    - Nemotron-Code-Metadata/part_00039.parquet59    - Nemotron-Code-Metadata/part_00040.parquet60    - Nemotron-Code-Metadata/part_00041.parquet61    - Nemotron-Code-Metadata/part_00042.parquet62    - Nemotron-Code-Metadata/part_00043.parquet63    - Nemotron-Code-Metadata/part_00044.parquet64    - Nemotron-Code-Metadata/part_00045.parquet65    - Nemotron-Code-Metadata/part_00046.parquet66    - Nemotron-Code-Metadata/part_00047.parquet67    - Nemotron-Code-Metadata/part_00048.parquet68    - Nemotron-Code-Metadata/part_00049.parquet69    - Nemotron-Code-Metadata/part_00050.parquet70    - Nemotron-Code-Metadata/part_00051.parquet71    - Nemotron-Code-Metadata/part_00052.parquet72    - Nemotron-Code-Metadata/part_00053.parquet73    - Nemotron-Code-Metadata/part_00054.parquet74    - Nemotron-Code-Metadata/part_00055.parquet75    - Nemotron-Code-Metadata/part_00056.parquet76    - Nemotron-Code-Metadata/part_00058.parquet77    - Nemotron-Code-Metadata/part_00059.parquet78    - Nemotron-Code-Metadata/part_00060.parquet79    - Nemotron-Code-Metadata/part_00061.parquet80    - Nemotron-Code-Metadata/part_00062.parquet81    - Nemotron-Code-Metadata/part_00063.parquet82    split: train83track_downloads: true84---85# Nemotron-Pretraining-Code-v386 87## Dataset Description:88 89The [Nemotron-Pretraining-Code-v3](https://huggingface.co/datasets/nvidia/Nemotron-Pretraining-Code-v3) dataset is part of the [Nemotron Pretraining Data](https://huggingface.co/collections/nvidia/nemotron-pre-training-datasets) collection of pretraining datasets. Designed for the [NVIDIA Nemotron 3](https://huggingface.co/collections/nvidia/nvidia-nemotron-v3) family of LLMs, this dataset is intended to improve the coding capabilities of LLMs.90 91The [Nemotron-Pretraining-Code-v3](https://huggingface.co/datasets/nvidia/Nemotron-Pretraining-Code-v3) dataset contains the metadata corresponding to the raw source-code update to our [Nemotron-Pretraining-Code-v2](https://huggingface.co/datasets/nvidia/Nemotron-Pretraining-Code-v2) and [Nemotron-Pretraining-Code-v1](https://huggingface.co/datasets/nvidia/Nemotron-Pretraining-Code-v1) source-code corpora. It adds 146 M new files (173 B tokens) from GitHub (with a cut-off date of September 30, 2025). **This dataset contains only incrementally new files and is meant to be used jointly with the v2 and v1 corpora.**92 93The dataset contains a single split Nemotron-Code-Metadata with the following columns:94 95- **commit_id:** The first seven characters of the Git commit hash96- **rel_path:** The file path of the content relative to the repository root directory97- **language:** The detected programming language of the content98 99This dataset is ready for commercial use.100 101## Dataset Owner(s):102 103NVIDIA Corporation104 105## Dataset Creation Date:106 10705/18/2026108 109## Version:110 111[Nemotron-Pretraining-Code-v3](https://huggingface.co/datasets/nvidia/Nemotron-Pretraining-Code-v3)112 113Previous Version(s):114- [Nemotron-Pretraining-Code-v2](https://huggingface.co/datasets/nvidia/Nemotron-Pretraining-Code-v2)115- [Nemotron-Pretraining-Code-v1](https://huggingface.co/datasets/nvidia/Nemotron-Pretraining-Code-v1)116 117Relationship to Previous Version(s): This is an **extension** of the previously released datasets. They are meant to be used together.118 119## License/Terms of Use:120 121The **Nemotron-Pretraining-Code-v3** dataset is licensed under the [Creative Commons Attribution 4.0 International License](https://creativecommons.org/licenses/by/4.0/) (CC-BY-4.0).122 123## Intended Usage:124 125The Nemotron-Pretaining-Code-v3 Dataset is intended to be used by the community to continue to improve open models.126 127## Dataset Characterization128 129Data Collection Method: Automated130 131Labeling Method: Automated132 133## Dataset Format134 135Modality: Text136 137Format: Parquet138 139## Dataset Quantification140 141Record Count: 146.3M samples142 143Measurement of Total Data Storage: 8.2 GB144 145## Reference(s):146 147If  you  use  our  dataset  in  your  research,  please  cite  our  [NVIDIA  Nemotron  3  Ultra  tech  report](https://research.nvidia.com/labs/nemotron/files/NVIDIA-Nemotron-3-Ultra-Technical-Report.pdf).148 149## Ethical Considerations:150 151NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. Developers should work with their internal developer teams to ensure this dataset meets requirements for the relevant industry and use case and addresses unforeseen product misuse.152 153Please report quality, risk, security vulnerabilities or NVIDIA AI Concerns [here](https://www.nvidia.com/en-us/support/submit-security-vulnerability/).