BatsResearch/bonito
A lightweight library for generating synthetic instruction tuning datasets for your data without GPT.
Bonito helps machine learning engineers and researchers quickly create new training data for instruction-tuned language models. You provide unannotated text, and Bonito generates relevant question-answer pairs or other task-specific examples. This is ideal for fine-tuning models on specialized tasks where labeled data is scarce or expensive to produce manually.
823 stars. No commits in the last 6 months.
Use this if you need to rapidly generate synthetic training datasets for custom instruction-tuning tasks without relying on large, external language models like GPT.
Not ideal if you require perfectly human-quality, nuanced annotations for extremely sensitive or safety-critical applications.
Stars
823
Forks
56
Language
Python
License
BSD-3-Clause
Category
Last pushed
Jul 15, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/BatsResearch/bonito"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
InternScience/GraphGen
GraphGen: Enhancing Supervised Fine-Tuning for LLMs with Knowledge-Driven Synthetic Data Generation
timothepearce/synda
A CLI for generating synthetic data
rasinmuhammed/misata
High-performance open-source synthetic data engine. Uses LLMs for schema design and vectorized...
ziegler-ingo/CRAFT
[TACL, EMNLP 2025 Oral] Code, datasets, and checkpoints for the paper "CRAFT Your Dataset:...
ZhuLinsen/FastDatasets
A powerful tool for creating high-quality training datasets for Large Language Models...