datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
terminalbench-sqlite-dbbird-critic-1.0-sqlite
📢 Update 2026-03-23
We release BIRD-Critic-SQLite, a dataset containing 500 high-quality user issues focused on real-world SQLite database applications. Along with the dataset, we also release three RL-trained models: BIRD-Talon-14B, BIRD-Talon-7B, and BIRD-Zeno-7B. The schema file is included in the code repository https://github.com/bird-bench/BIRD-CRITIC-1/blob/main/baseline/data/sqlite_schema.jsonl
BIRD-CRITIC-1.0-SQLite
BIRD-Critic is the first SQL debugging… See the full description on the dataset page: https://huggingface.co/datasets/birdsql/bird-critic-1.0-sqlite.cbdb-sqliteYou can download the newest CBDB SQLite database here
You can download the historical CBDB SQLite database here
livesqlbench-base-lite-sqlite
🚀 LiveSQLBench-Base-Lite
A dynamic, contamination‑free benchmark for evaluating LLMs on complex, real‑world text‑to‑SQL tasks.
🌐 LiveSQLBench Website • 🌐 BIRD-INTERACT Project Page • 📄 Paper • 💻 LiveSQLBench GitHub • 💻 BIRD-INTERACT GitHub
Maintained by the 🦜 BIRD Team @ HKU & ☁️ Google Cloud
📊 LiveSQLBench Overview
LiveSQLBench (BIRD-SQL Pro v0.5) is a contamination-free, continuously evolving benchmark designed to evaluate LLMs on complex, real-world… See the full description on the dataset page: https://huggingface.co/datasets/birdsql/livesqlbench-base-lite-sqlite.six-gym-sqlite
📢 Update 2026-03-23
We release BIRD-Critic-SQLite, a dataset containing 500 high-quality user issues focused on real-world SQLite database applications. This dataset is the train split of BIRD-Critic-SQLite, comprising 5,000 data instances for model training and development. Along with the dataset, we also release three RL-trained models: BIRD-Talon-14B, BIRD-Talon-7B, and BIRD-Zeno-7B.
📋 Dataset Structure
Below is a description of the dataset fields and additional… See the full description on the dataset page: https://huggingface.co/datasets/birdsql/six-gym-sqlite.conceptnet-sqlitePlease see here for related source code for reading the files: https://github.com/coldstore/conceptnet-sqlite
A newer and better version of this is at: https://github.com/ysenarath/kblite
Danbooru2021-SQLite
Danbooru 2021 SQLite
Dataset Summary
This is the metadata of danbooru 2021 dataset in SQLite format.
https://gwern.net/danbooru2021
Supported Tasks and Leaderboards
[More Information Needed]
Languages
[More Information Needed]
Dataset Structure
Data Instances
[More Information Needed]
Data Fields
[More Information Needed]
Data Splits
[More Information Needed]
Dataset Creation
Curation… See the full description on the dataset page: https://huggingface.co/datasets/KBlueLeaf/Danbooru2021-SQLite.recipes-sqlite-dben-wiktionary-sqlite-full
English Wiktionary - Normalized SQLite Database
This is a normalized SQLite database of English Wiktionary, capturing every field from the cstr/en-wiktionary-extracted dataset.
Note that this does not include all that would be extractable per wiktextract with --all, like translations, examples, etc.
🎯 Key Features
fields captured including:
🔗 Wikilinks in definitions (semantic connections)
📝 Qualifiers (e.g., "archaic", "US", "informal")
🏷️ Sense IDs (unique… See the full description on the dataset page: https://huggingface.co/datasets/cstr/en-wiktionary-sqlite-full.gemma-4-e2b-SAE-sqlite
Gemma 4 E2B SAE SQLite Atlas
An exact, queryable SQLite representation of all 35 residual-stream
sparse autoencoders from
juiceb0xc0de/gemma-4-e2b-it-SAE.
The database contains 1,720,320 feature rows. Encoder and decoder
vectors preserve the source checkpoints' float32 values exactly.
Files
gemma-4-e2b-sae.sqlite3 — SQLite database (20.63 GiB)
manifest.json — source revision, dimensions, SHA-256, and integrity result
Database SHA-256:… See the full description on the dataset page: https://huggingface.co/datasets/juiceb0xc0de/gemma-4-e2b-SAE-sqlite.SOONibus_Theory_for_RAG_chroma_sqlite3
THEORY SOONibus (#1) for Chroma
This is a database of texts held in the form of vector embeddings
Check out the JSONL version HERE for training/direct access!
An eclectic selection from the library of books, articles, papers, and varied textual curios I've amassed over the years.
Contains works in English, Russian, and French, including a wealth of critical theory, philosophy, translation theory, comparative literature, literary ethics, radical/revolutionary politics (mainly… See the full description on the dataset page: https://huggingface.co/datasets/AlekseyCalvin/SOONibus_Theory_for_RAG_chroma_sqlite3.danbooru-2023-sqlite-fixed-7110548
SQLITE-DB for Danbooru 2023 (until 7110548)
This is an cleaned-up version (almost totally recreated) of https://huggingface.co/datasets/KBlueLeaf/danbooru2023-sqlite
The previous sqlite database had major defects, especially with tag ids being mismatched, which was causing data to be actually different from server.
Note that minor information, such as uploader id, are not fixed.
Most of the discrepancy has been detected from id 5139963-6859952.
The additional scripts and example… See the full description on the dataset page: https://huggingface.co/datasets/AngelBottomless/danbooru-2023-sqlite-fixed-7110548.zsql-sqlite-dpo
zsql-sqlite-dpo
This is a dataset for training machine learning models to convert natural
English language text into SQLite dialect SQL queries.
This dataset comprises 200,000 DPO pairs curated to support the rapid
development of text-to-SQL generation models. The uniqueness of this dataset
lies in its optimization process. The "chosen" field within each data pair
contains SQL queries that have been canonicalized, optimized, and which are
chosen from the candidate set which… See the full description on the dataset page: https://huggingface.co/datasets/zerolink/zsql-sqlite-dpo.ontology-sqlite-vectorstore
Ontology & Embedding Database
This repository provides the local ontology database to be used by StructSense (in general can be used for any other purpose), including both structured ontological data and precomputed vector embeddings for efficient semantic search.
Overview
3,925,124 classes
2,683,756 synonyms
Data storage:
SQLite → structured ontology data (bioportal.db)
Embeddings → vector embeddings (to be uploaded)
Vector embeddings (includes preferred label +… See the full description on the dataset page: https://huggingface.co/datasets/sensein/ontology-sqlite-vectorstore.de-wiktionary-sqlite-normalized
German Wiktionary - Normalized SQLite Database
A fully normalized, production-ready SQLite database of German Wiktionary with complete linguistic information and optimized query performance.
🎯 Key Features
✅ Zero data loss: All information from original Wiktionary preserved
⚡ Lightning-fast queries: Comprehensive indexing (< 5ms typical queries)
🔍 Full grammatical analysis: Complete inflection paradigms, word forms, 185 unique grammatical tags
🔗 Semantic relations:… See the full description on the dataset page: https://huggingface.co/datasets/cstr/de-wiktionary-sqlite-normalized.Gelbooru-SQLiteThis is db dump(s) of booru-typed databases.
The codebase (https://github.com/aria1th/Booru-Unified-Sqlite) will be used for creating DB, to handle various types of DB + allowing multiple DBs being loaded in same program.
Danbooru DB, mainly, will be updated at https://huggingface.co/datasets/KBlueLeaf/danbooru2023-sqlite too.
SQLite-database_Query_Demo_Script# Chinook_Sqlite.sqlite SQLite database and SQL Query Demo with Tkinter
No API keys are required.
Executables:
Demonstrate_SQL_Queries_and_Display_results_in_Chinook_Sqlite_CMD_console_and_in_Popoup_Tkinter.py
download_direct_SQL_dbs.py # for obtaining the included Chinook_Sqlite.sqlite from where I downloaded it on April 10, 2025
Both of these include def_function_to_generate_installed_packages_list_file
Data:
Chinook_Sqlite.sqlite
the download_direct_SQL_dbs.py script also… See the full description on the dataset page: https://huggingface.co/datasets/MartialTerran/SQLite-database_Query_Demo_Script.de-wiktionary-sqlite-full
German Wiktionary - FULL Normalized SQLite Database
This is a complete, lossless, and fully normalized SQLite database of German Wiktionary, capturing 100% of the structured data from the cstr/de-wiktionary-extracted dataset.
It is designed for production-ready applications, complex linguistic analysis, and mobile apps (Flutter, React Native) that require a comprehensive local dictionary.
🎯 Key Features
✅ 100% Lossless: All 30+ top-level and nested fields from the… See the full description on the dataset page: https://huggingface.co/datasets/cstr/de-wiktionary-sqlite-full.Kaikki-Wiktionary-Ultimate-SQLite
Kaikki Ultimate Raw SQLite - World Dictionary Database (2026 Edition)
📌 Overview
This dataset is a high-performance SQLite conversion of the massive Kaikki.org (Wiktextract) raw data. It contains millions of lexical entries across thousands of languages, preserved in its absolute raw JSON format to ensure zero data loss.
This database is designed for developers, linguists, and AI researchers who need a structured, indexed, and offline-ready version of the world's most… See the full description on the dataset page: https://huggingface.co/datasets/wave101828228/Kaikki-Wiktionary-Ultimate-SQLite.rampdb-sqliteen-wiktionary-sqlite-all
English Wiktionary - TRULY LOSSLESS Normalized SQLite Database
This is a 100% lossless, fully normalized SQLite database of English Wiktionary, capturing EVERY field from the cstr/en-wiktionary-extracted-all dataset.
🎯 Key Features
✅ 100% Lossless: ALL fields captured including:
🔗 Wikilinks in definitions (semantic connections)
📝 Qualifiers (e.g., "archaic", "US", "informal")
🏷️ Sense IDs (unique identifiers)
🌐 Wikidata IDs (for semantic web linking)
📚 Attestations… See the full description on the dataset page: https://huggingface.co/datasets/cstr/en-wiktionary-sqlite-all.gutenburg-sqlitebird-sqlite-sft-train
BIRD SQLite Text-to-SQL SFT Dataset
Supervised fine-tuning data for a SQLite-dialect text-to-SQL specialist model,
built from the BIRD benchmark train split.
Contents
7,483 train + 408 val examples spanning 69 distinct database schemas
(movie_platform, chicago_crime, hockey, mondial_geo, works_cycles, and 64
others), split by a stratified per-database 95/5 hold-out (sft_sqlite_ train.jsonl / sft_sqlite_val.jsonl) with zero exact overlap between them.
Format:… See the full description on the dataset page: https://huggingface.co/datasets/hiimivantang/bird-sqlite-sft-train.spider2_lite_sqlite_summerzsqlite-vec-ann-demosde-wiktionary-sqlite
German Wiktionary SQLite Database
This is a normalized SQLite database version of the German Wiktionary data.
Source
Original data from: cstr/de-wiktionary-extracted
Database Schema
The database is normalized into the following tables:
entries: Main word entries (word, pos, lang, etc.)
senses: Word definitions and meanings
translations: Translations to other languages
sounds: Pronunciation information
synonyms: Synonymous words
tags: Tags/labels (many-to-many… See the full description on the dataset page: https://huggingface.co/datasets/cstr/de-wiktionary-sqlite.dewiki-sqlite-zstd-fts5
German Wikipedia Compressed SQLite Database (FTS5 + Shared Zstd Frames)
A heavily compressed, fully queryable SQLite database containing the complete German Wikipedia (~2.66M articles). Designed specifically for offline-first RAG applications, local LLMs, and resource-constrained edge devices (Android, iOS, Raspberry Pi).
💡 Key Specifications
Articles: 2,660,620 articles in 13,504 shared zstd chunks.
Size: ~4.83 GB single .db file.
Full-Text Search: SQLite FTS5… See the full description on the dataset page: https://huggingface.co/datasets/sauerkrieger/dewiki-sqlite-zstd-fts5.knowledgebase-filtered-sqlite
Knowledgebase Filtered (SQLite)
College-math knowledge graph as a queryable SQLite + FTS5 database, designed
for retrieval / tool-call agents (kg_search_statement, kg_get_statement,
kg_get_dependencies, md_fetch_section, ...).
What's filtered
The full corpus (887,223 statement nodes) is retained so dependency lookups and
full-text search stay complete. Two annotation columns were added to mark the
cleaned anchor pool — statements good enough to seed an exam… See the full description on the dataset page: https://huggingface.co/datasets/blackBOX25I47/knowledgebase-filtered-sqlite.sqli-test-httpsqwen25-coder-7b-instruct_sqlite