Emoji API

The Emoji API bundles approximately 1,914 emoji directly into the edge Worker. Each one comes enriched with its character, human-readable name, URL-friendly slug, Unicode group, a skin-tone support flag, the Unicode version it was introduced in, and precomputed codepoints. The dataset spans all standard Unicode groups including Smileys & Emotion, People & Body, Animals & Nature, Food & Drink, Travel & Places, Activities, Objects, Symbols, and Flags.

Call GET /emoji/v1/search with a q parameter. The match is case-insensitive and partial, applied against both the emoji name and its slug — so q=grin will return grinning_face, grinning_face_with_big_eyes, and more. You can optionally scope results to a single Unicode group with the group parameter and cap the number of results with limit (1-200, default 50). An empty q, an invalid limit, or an unknown group returns a 400 error.

The GET /emoji/v1/lookup endpoint resolves the full record for a single emoji. You can provide either the exact emoji character via the emoji parameter (URL-encoded) or a slug like grinning_face via the slug parameter. At least one is required. If you supply both, the emoji character takes precedence. If neither is provided you get a 400, and if no emoji matches you get a 404 with an explicit error message.

GET /emoji/v1/random returns random emoji using the count parameter (1-50, default 1). Sampling is done without replacement, so you will never get duplicate emoji within a single response, and the selection varies on every call. You can also pass an optional case-insensitive group filter to draw only from a specific category, such as group=Food & Drink. A count outside the 1-50 range or an unknown group returns a 400 error.

The codepoints field is an array of Unicode codepoint strings in U+XXXX form that make up the emoji. Many emoji are not a single codepoint — they may combine a base character with a variation selector (like U+FE0F) or join multiple codepoints with a zero-width joiner (ZWJ). The API precomputes and correctly expands these sequences so you can reconstruct, analyze, or compare emoji precisely without parsing the raw string yourself.

The entire emoji dataset is bundled into a Cloudflare Worker and served from memory at the edge, so there are no upstream calls or third-party hops. Typical compute time is well under a millisecond, and latency is dominated by the network round-trip to the nearest of Cloudflare's 300+ edge locations. This makes the API ideal for high-throughput autocomplete and latency-sensitive UIs. Authentication is a simple apikey query parameter on every request.