Synthetic Data Generation LLM Tools

Tools for generating synthetic datasets and training data for LLMs through various methods (QA pairs, tabular data, code, structured extraction). Does NOT include general data processing, data augmentation for images, or dataset annotation/curation platforms.

There are 42 synthetic data generation tools tracked. 3 score above 50 (established tier). The highest-rated is InternScience/GraphGen at 56/100 with 978 stars. 1 of the top 10 are actively maintained.

Get all 42 projects as JSON

curl "https://pt-edge.onrender.com/api/v1/datasets/quality?domain=llm-tools&subcategory=synthetic-data-generation&limit=20"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.

#	Tool	Score	Tier	Stars	Language
1	InternScience/GraphGen GraphGen: Enhancing Supervised Fine-Tuning for LLMs with Knowledge-Driven...	56	Established	978	Python
2	rasinmuhammed/misata High-performance open-source synthetic data engine. Uses LLMs for schema...	53	Established	52	Python
3	timothepearce/synda A CLI for generating synthetic data	50	Established	43	Python
4	dmanuel64/codablellm A framework for creating and curating high-quality code datasets tailored...	39	Emerging	3	Python
5	ZhuLinsen/FastDatasets A powerful tool for creating high-quality training datasets for Large...	39	Emerging	183	Python
6	ziegler-ingo/CRAFT [TACL, EMNLP 2025 Oral] Code, datasets, and checkpoints for the paper "CRAFT...	39	Emerging	34	Python
7	BatsResearch/bonito A lightweight library for generating synthetic instruction tuning datasets...	36	Emerging	823	Python
8	Oqura-ai/deepresearch-datagen-cli Using deep research workflow to generate datasets for finetuning LLMs.	36	Emerging	39	Python
9	Alannikos/edg4llm A unified tool to generate fine-tuning datasets for LLMs, including...	34	Emerging	62	Python
10	asaparov/prontoqa Synthetic question-answering dataset to formally analyze the...	34	Emerging	156	Python
11	nalinrajendran/synthetic-LLM-QA-dataset-generator Create synthetic datasets for training and testing Language Learning Models...	30	Emerging	54	Python
12	Itachi-Uchiha581/Auto-Data Auto Data is a library designed for quick and effortless creation of...	29	Experimental	106	Python
13	ISE-FIZKarlsruhe/concept_extraction ConExion	28	Experimental	6	Python
14	GURPREETKAURJETHRA/Synthetic-Data-Generation-using-LLM Synthetic Data Generation using LLM via Argilla, Distilabel, ChatGPT, etc.	27	Experimental	5	Jupyter Notebook
15	kevinscaria/TarGEN Targeted Data Generation with Large Language Models	27	Experimental	19	Jupyter Notebook
16	Glavin001/Data2AITextbook 🚀 Automatically convert unstructured data into a high-quality 'textbook'...	26	Experimental	25	Jupyter Notebook
17	BothBosu/Synthetic-Data-for-Scam-Detection-Leveraging-LLMs-to-Train-Deep-Learning-Models This repository contains the source code and synthetic datasets used in the...	25	Experimental	6	Jupyter Notebook
18	copyleftdev/faux-foundry FauxFoundry - Synthetic data generation powered by local LLMs	25	Experimental	2	Go
19	danmurf/datakeg Brew synthetic training data from your documentation using LLMs	22	Experimental	—	Go
20	Red1998/faux-foundry 🤖 Generate unique synthetic datasets effortlessly with FauxFoundry, using...	22	Experimental	—	Go
21	dmeldrum6/synthetic-dataset Web based tool for generating Q&A datasets from an LLM	22	Experimental	—	HTML
22	jehumtine/synthetic_data_generator This script is designed to convert bodies of text into a question and answer...	20	Experimental	24	Python
23	rodrigobnogueira/faker-ai-provider 🤖 Faker provider for generating AI/ML fake data - models, companies,...	20	Experimental	1	Python
24	MelNajkar/llm-data-augmentation-sentiment LLM-based synthetic data generation for improving sentiment classification...	19	Experimental	—	Jupyter Notebook
25	jqwangai/SynPT An Improved Data Synthesis Method Driven by Large Language Models for...	18	Experimental	3	Python
26	Pro-GenAI/DataClassifier An AI-driven approach to Label LLM Training Data	18	Experimental	4	Python
27	yzhan238/TELEClass The source code used for paper "TELEClass: Taxonomy Enrichment and...	17	Experimental	25	Python
28	nphdang/Pred-LLM Generating tabular data via Large Language Models (LLMs)	15	Experimental	6	Python
29	aekpalakorn/TGRE-Classification Source code for the taxonomic knowledge assessment and occupation/skill...	15	Experimental	—	Python
30	ScottishCoder/AuldLangSynth AuldLangSynth is an open-source data-centric language synthesis platform...	15	Experimental	—	JavaScript
31	MichiganNLP/depression_synthetic_data Can LMs generate useful synthetic data for the mental health domain?	15	Experimental	6	Jupyter Notebook
32	Chessperson/multiomics-synth R synthpop for proteomics/metabolomics cohorts (your 30 cohorts, 6k+ cols)....	14	Experimental	—	R
33	ChandanKSahu/ReqList_ReqNet_ReqSim ReqList, ReqNet and ReqSim Datasets for the publication `A Network and...	14	Experimental	7	HTML
34	jd-coderepos/scisynthesis for prompts, dataset, and code addressing the task of scientific synthesis	14	Experimental	4	Python
35	ZEKE320/llm-dataset-generator The LLM Dataset Generator is an open source tool for generating text data...	13	Experimental	6	Jupyter Notebook
36	pezzos/jsonl_dataset_generator Generate rich JSONL datasets from topics to fine-tune Large Language Models....	13	Experimental	—	JavaScript
37	CartographerLabs/Lights-Camera-Extremism A Social Network Synthetic Dataset Generation Framework	12	Experimental	3	Python
38	tiddly-gittly/TiddlyWiki-LLM-dataset WikiText syntax dataset generation pipeline and open dataset for auto UI...	12	Experimental	3	TypeScript
39	daspartho/DistillClassifier Easily generate synthetic data for classification tasks using LLMs	11	Experimental	2	Python
40	AikyamLab/regtext A framework to generate unlearnable text data	11	Experimental	2	Python
41	Ki-Seki/autotab Automatically fill in missing values in tabular data using in-context...	10	Experimental	1	Python
42	thalesbertaglia/instasynth Synthetic Instagram Post Generation for Social Media Research	10	Experimental	1	Jupyter Notebook

Comparisons in this category

synthetic-LLM-QA-dataset-generator and synthetic-dataset (30 vs 22)