Instruction Tuning Python Packages

llamafactory

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

convmerge

Merge heterogeneous chat/text sources into a single LLM training format (JSONL)

bespokelabs-curator

Synthetic data curation for post-training and structured data extraction

py-data-juicer

Data processing for and with foundation models! 🍎 🍋 🌽 ➡️ ➡️🍸 🍹 🍷

gritlm

Generative Representational Instruction Tuning

tasksource

Datasets collection and preprocessings framework for NLP extreme multitask learning

datadreamer-dev

DataDreamer: Prompt. Generate Synthetic Data. Train & Align Models.   🤖💤

llmtuner

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

decontaminate

`decon`, but with python API binding.

lazyllm-llamafactory

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

scar-tool

[ACL 2025 main] SCAR: Data Selection via Style Consistency-Aware Response Ranking for Efficient Instruction-Tuning of Large Language Models

otter-ai

🦦 Otter, a multi-modal model based on OpenFlamingo (open-sourced version of DeepMind's Flamingo), trained on MIMIC-IT and showcasing improved instruction-following and in-context learning ability.

simplifine-alpha

🚀 Easy, open-source LLM finetuning with one-line commands, seamless cloud integration, and popular optimization frameworks. ✨

llava-torch

[NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond.

vibeprompt

🦩VibePrompt. A lightweight Python package for adapting prompts by tone, style, and audience. Built on top of LangChain, VibePrompt supports multiple LLM providers and enables structured, customizable prompt transformations for developers, writers, and researchers.

castwright

Generate synthetic instruction-tuning data from seed examples. Simple API, built-in quality filtering, multi-provider.

llamafactory-songlab

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)