datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Roblox-luau-coding_L1
8BitStudio/Roblox-luau-coding_L1
A dataset for training and fine-tuning AI models on Roblox Luau scripting.
Covers a wide range of scripting topics from beginner to advanced.
Summary
This dataset contains 12,306 Luau code examples designed to teach AI models
how to write scripts for Roblox. Topics range from basic part manipulation
to complex datastore systems.
Dataset Structure
Data Format
Each example is a tab-separated pair of a… See the full description on the dataset page: https://huggingface.co/datasets/8BitStudio/Roblox-luau-coding_L1.Luau-Coder-1.0-Preview-SFT
Luau Coder 1.0 Preview SFT 🦭
This dataset is exceptionally high-quality supervised fine-tuning conversations for a highly capable coding model in Roblox Luau domain.
It prioritize technical correctness, useful engineering judgment, realistic interaction, and efficient explanations over output volume.
This dataset includes & covering:
Multi-turns (4-10 turns)
Dynamic CoT (length)
Dynamic Interleaved Reasoning
Long Context Session
Q/A
Review
Debugging
Bug Fix… See the full description on the dataset page: https://huggingface.co/datasets/khtsly/Luau-Coder-1.0-Preview-SFT.luau-stack-hq
[!Note]
Last collected: 2026-09-05 23:33
Luau Stack HQ 🦭
Luau Stack HQ is a highly curated & deduplicated Luau source code with Type-script included from github repository with linter (stylua), it contains ~18.11M tokens (17903 unique files, 16313 luau files & 1590 ts files). It is built strictly for the training stages, providing a pristine density of type-safe, functional architecture without the noise of standard web scrapes.
# Filtering Pipeline
To ensure… See the full description on the dataset page: https://huggingface.co/datasets/khtsly/luau-stack-hq.roblox-luau-dataset
Roblox Luau Instruct Dataset
This dataset contains 565,760 lines of Roblox Luau code, cleaned of duplicates and malicious code.
It was used to refine the language model to improve coding skills in Roblox.
Statistik
Line: 565,760
Format: JSONL (Instruction, Input, Output)
File Size: 1.6 GB
Empathetic_counseling_Dataset
Dataset Card for Dataset Name
Empathetic_counseling is a dataset intended for training conversational language models for generating text in empathetic and mental counseling dialogues.
This dataset card aims to be a base template for new datasets. It has been generated using this raw template.
Dataset Details
Dataset Description
This is a dataset resulting after concatenating some examples from the "empathetic_dialogues" dataset with a dataset resulting… See the full description on the dataset page: https://huggingface.co/datasets/LuangMV97/Empathetic_counseling_Dataset.luanti-complete-dataset
Complete Luanti Package Dataset
Description
Complete Luanti Package Dataset for Luanti (Minetest) expertise fine-tuning.
Dataset Information
Size: 2,592 entries
Format: Harmony format for LLM fine-tuning
Source: Luanti ContentDB package collection
Quality: Filtered and validated Luanti package metadata
Usage
from datasets import load_dataset
dataset = load_dataset("ToddLLM/luanti-complete-dataset")
print(dataset)
Schema
Each entry… See the full description on the dataset page: https://huggingface.co/datasets/ToddLLM/luanti-complete-dataset.luau-org-web-textluau-repo-docs-text
[!Note]
Last collected: 2026-09-05 23:51
Contains 306 READMEs & 1,544 Doc pages (~1.79M tokens)
luaucoder-instructions-v3-SFTthis dataset includes 12500 multi-turn input reasoning and output pairs of strictly only Roblox luau related code generation tasks, keep in mind this is 100% synthetic data and may need filtering.
original prompts based on: TorpedoSoftware/Roblox-Luau-Reasoning-v1.0
For questions, https://huggingface.co/datasets/Pinkstack/luaucoder-instructions-v3-SFT/discussions
Roblox_Luau_CoT_conversational_sharegpt_lqv1New version soon
This is a dataset based on Roblox/luau_corpus, with a sharegpt style, modified Thought and Output tokens, with a proper conversational style.
This highly experimental dataset is designed to help SLMs and LLMs handle reasoning with Luau Roblox code generation, it has the same style of tokens as Openo1
Ideally, after tuning your LLm with the Roblox/luau_corpus dataset, fine-tune it with another dataset like this one to create LLms similar to superthoughts by us, openo1, deepseek… See the full description on the dataset page: https://huggingface.co/datasets/Pinkstack/Roblox_Luau_CoT_conversational_sharegpt_lqv1.psct_lua
YGOPro Lua Code Generation Dataset
Dataset Description
This dataset contains Yu-Gi-Oh! card effects written with correct PSCT (Problem-Solving card text) paired with their corresponding YGOPro Lua script implementations. It's designed for training models to generate functional Lua code for YGOPro (Yu-Gi-Oh! Pro) simulator from natural language card effect descriptions.
Dataset Structure
Data Fields
instruction: The task instruction (constant across… See the full description on the dataset page: https://huggingface.co/datasets/lenarc/psct_lua.lua-manuals
Lua Programming Language Documentation
This dataset contains the Lua programming language documentation,
chunked using semantic parsing for pretraining language models.
Updated: 2025-09-08
Loading
from datasets import load_dataset
ds = load_dataset("json", data_files={"train": "train.jsonl"}, split="train")
Statistics
Format: JSONL with single text field per line
Chunking: Semantic structure-aware chunking
Content: Official Lua documentation and manuals
the-luau-stack
The Luau Stack
The Luau Stack is a collection of open sourced code from GitHub, formatted with StyLua. Any use of all or part of the code gathered in The Luau Stack must abide by the terms of the original licenses, including attribution clauses when relevant. We facilitate this by providing provenance information for each data point.
The Luau Stack may be updated to enact validated data removal requests. You agree to keep your copies up to date with The Luau Stack to abide by data… See the full description on the dataset page: https://huggingface.co/datasets/TorpedoSoftware/the-luau-stack.luau-stack-hq-ChatML
Dataset Card for YCWTG/luau-stack-hq-ChatML
Summary
This dataset is a pretraining-oriented repackaging of khtsly/luau-stack-hq.
It keeps only the source code content and stores it in a ChatML-style messages field.
Rows: 21,704
File format: JSONL
Split: train
Main file: train.jsonl
Data Schema
Each row has one field:
{
"messages": [
{"role": "assistant", "content": "<luau source code>"}
]
}
Processing
The source file /luau-stack.jsonl… See the full description on the dataset page: https://huggingface.co/datasets/YCWTG/luau-stack-hq-ChatML.luanti-copyleft-dataset
Copyleft Luanti Package Dataset
Description
Copyleft Luanti Package Dataset for Luanti (Minetest) expertise fine-tuning.
Dataset Information
Size: 1,073 entries
Format: Harmony format for LLM fine-tuning
Source: Luanti ContentDB package collection
Quality: Filtered and validated Luanti package metadata
Usage
from datasets import load_dataset
dataset = load_dataset("ToddLLM/luanti-copyleft-dataset")
print(dataset)
Schema
Each entry… See the full description on the dataset page: https://huggingface.co/datasets/ToddLLM/luanti-copyleft-dataset.W3-NewsAlpha
W³: A News-to-Alpha Benchmark for Utility-Driven Web3 Recommendation and Trading Backtesting
Source platform: CoinMeta (Bijie, a leading Web3 information aggregator)
Time span: January 1, 2022 — January 1, 2026 (4 years)
TL;DR
W³ is the first benchmark dedicated to Web3 news recommendation that goes beyond ranking accuracy. It links a real-world industrial-grade news platform to tick-level cryptocurrency market data and a realistic trading backtesting protocol… See the full description on the dataset page: https://huggingface.co/datasets/jining-luan/W3-NewsAlpha.luanti-commercial-dataset
Commercial-Safe Luanti Package Dataset
Description
Commercial-Safe Luanti Package Dataset for Luanti (Minetest) expertise fine-tuning.
Dataset Information
Size: 1,387 entries
Format: Harmony format for LLM fine-tuning
Source: Luanti ContentDB package collection
Quality: Filtered and validated Luanti package metadata
Usage
from datasets import load_dataset
dataset = load_dataset("ToddLLM/luanti-commercial-dataset")
print(dataset)
Schema… See the full description on the dataset page: https://huggingface.co/datasets/ToddLLM/luanti-commercial-dataset.luanti-capability-eval
Luanti Capability Evaluation Dataset
Description
Luanti Capability Evaluation Dataset for Luanti (Minetest) expertise fine-tuning.
Dataset Information
Size: 60 entries
Format: Harmony format for LLM fine-tuning
Source: Luanti ContentDB package collection
Quality: Filtered and validated Luanti package metadata
Usage
from datasets import load_dataset
dataset = load_dataset("ToddLLM/luanti-capability-eval")
print(dataset)
Schema
Each… See the full description on the dataset page: https://huggingface.co/datasets/ToddLLM/luanti-capability-eval.luanti-capability-training
Luanti Capability Training Dataset
Description
Luanti Capability Training Dataset for Luanti (Minetest) expertise fine-tuning.
Dataset Information
Size: 600 entries
Format: Harmony format for LLM fine-tuning
Source: Luanti ContentDB package collection
Quality: Filtered and validated Luanti package metadata
Usage
from datasets import load_dataset
dataset = load_dataset("ToddLLM/luanti-capability-training")
print(dataset)
Schema
Each… See the full description on the dataset page: https://huggingface.co/datasets/ToddLLM/luanti-capability-training.