About Emoji Dictionary
An open emoji dictionary built from the Unicode Standard and CLDR as deterministic primary sources, balancing technical accuracy with practical explanations.
Principles and design
- Deterministic standards data: Emoji entries, code point sequences, groups, and CLDR names and keywords are generated deterministically from official Unicode data (Unicode — / Emoji — / CLDR —).
- Limited LLM use: Language models such as OpenAI are used only for editorial content such as meaning, context, nuance, examples, and search aliases. They never generate the emoji inventory itself.
- RGI Emoji Sequence entries: Entries are RGI emoji sequences rather than only single code points, including ZWJ sequences, skin tone modifiers, and flags.
- Reproducibility and integrity: SHA-256 hashes are recorded for all downloaded source files so version updates can be synchronized safely and idempotently.
Primary data sources
| Unicode Standard | Version — |
|---|---|
| Emoji Specification | Emoji — (0 RGI sequences) |
| CLDR Version | CLDR — (Annotations & Derived Annotations: ja, en) |
| Official source files | emoji-test.txt, emoji-sequences.txt, emoji-zwj-sequences.txt, emoji-data.txt, emoji-variation-sequences.txt, UnicodeData.txt, DerivedAge.txt |
Attribution, license, and disclaimer
Unicode® Terms of Use / License Notice
Copyright © 1991-2026 Unicode, Inc. All rights reserved. Distributed under the Terms of Use in https://www.unicode.org/copyright.html .
Vendor emoji images
This site does not collect or redistribute proprietary emoji artwork owned by vendors such as Apple, Google, Microsoft, or Samsung. Emoji are rendered by the system fonts available on your operating system and browser.
System architecture
- Authoring and seed database: Local SQLite (
database/emoji.sqlite3) - Production database: Cloudflare D1 (distributed edge SQL database)
- Search engine: SQLite FTS5 (Prefix Search + Trigram Tokenizer)
- Frontend: Nuxt 3 (Vue 3, Nitro SSR on Cloudflare Workers)