Chinese variant-character lookup. Enter any Chinese character to see its semantic, simplified, traditional, Z-, specialized and spoofing variants from the Unicode Unihan database. Runs locally in your browser.
Chinese Variant Character Lookup (異體字)
How to use
Enter character(s)
Type one or more Chinese characters into the box — separated by spaces, or just paste a sentence. The tool extracts the Chinese characters and de-duplicates them automatically.
Look up variants
Click "Look Up Variants". Each character gets a card listing its variants grouped by type: semantic (異體字), simplified, traditional, Z-variant, specialized and spoofing.
Click a variant to chase it
Every variant in the results is a clickable button. Tap one to drop it into the input and look up its variants, so you can trace a chain of related forms.
Read the categories
Each group maps to a Unihan database field: semantic variant, simplified/traditional pair, Z-variant (a unifiable shape variant), specialized semantic variant, and spoofing (visual look-alike).
Variant Characters: One Word, Many Shapes
Across thousands of years of evolution, Chinese characters frequently developed "the same word, written differently." These characters — identical in meaning and pronunciation but different in written form — are called variant characters (異體字, yìtǐzì). For example 群 / 羣, 峰 / 峯, and 裡 / 裏 are each a single word with more than one accepted shape. Variants arise for many reasons: components swapping between top-bottom and left-right arrangements, simplified and full-stroke forms coexisting, and centuries of differences accumulated through hand-copying, woodblock printing, name-taboo avoidance, and regional usage. This tool reads the variant-relationship fields directly from the Unicode Unihan database and lists, in one place, every recorded variant of a character so you can compare forms quickly.
What the six categories mean
Unihan records different kinds of shape relationships in different fields, which this tool groups into six categories. A semantic variant (異體字, kSemanticVariant) is an alternate writing whose meaning is interchangeable. Simplified (kSimplifiedVariant) and traditional (kTraditionalVariant) forms are the simplification correspondences introduced from 1956 onward, such as 国↔國 and 变↔變. A Z-variant (kZVariant) is a shape that differs so slightly it could have been unified in Unicode but was encoded separately because it came from different sources. A specialized variant (kSpecializedSemanticVariant) is interchangeable only under a specific sense. And a spoofing variant (kSpoofingVariant) is a visual look-alike that could be used for deception, such as domain-name spoofing. Knowing these distinctions is what lets you judge whether two characters are genuinely interchangeable.
Who uses variant-character lookup
Variant lookup is useful in several settings. Classical-text editing and collation requires recognising the colloquial variants found in old printed editions. Name and document handling frequently runs into "same name, different character" issues across traditional/simplified and variant forms — especially in Hong Kong, Macau, Taiwan, and overseas Chinese communities. Typesetting and font development needs to confirm whether a glyph has a standard counterpart. And information security cares about the homograph-attack risk posed by spoofing variants and Z-variants. One important caveat: Unihan records objective shape correspondences, not orthographic recommendations — the fact that a character has a database variant does not mean you may freely substitute it in your specific context (for example simplified-Chinese standards or textbook usage). Judge against the prevailing standard and your actual purpose. All lookups in this tool run locally in your browser; nothing is uploaded or stored.
"A variant character is not a misspelling — it is the same word leaving different footprints across history and geography."
About the data source
Every variant relationship in this tool comes from the Unicode® Unihan database (Unihan_Variants.txt) — the authoritative character-property dataset maintained by the Unicode Consortium, covering tens of thousands of Han characters and used worldwide by operating systems, fonts, and input methods. It is contributed jointly by the national standards bodies of China, Japan, Korea, Vietnam and others, so a single character may carry several variants drawn from different sources at once. The data is an objective code-point correspondence; this tool only displays it locally, adding nothing and making no subjective ruling on which form is "correct." For publishing or formal documents, cross-check against the current national or regional orthographic standard.
10 Facts about Variant Characters
A variant character has the same meaning and reading but a different written form, e.g. 群/羣 or 峰/峯. They are the same word — not misspellings.
All variant data here comes from the Unicode Unihan database — the authoritative Han-character property dataset shared by operating systems and fonts worldwide.
Simplified and traditional forms are themselves a kind of variant relationship: 国↔國 and 变↔變 are the correspondences recorded in the kSimplifiedVariant and kTraditionalVariant fields.
A Z-variant (kZVariant) is a near-identical shape that could have been unified in Unicode but was encoded separately because of different sources — the difference is often hard to see.
A spoofing variant (kSpoofingVariant) is a confusable look-alike; security teams watch it for homograph attacks such as look-alike domain-name spoofing.
One character can carry several kinds of variant at once. Unihan is contributed jointly by the standards bodies of China, Japan, Korea, Vietnam and others, so different sources yield different records.
Variants are not always substitutable: Unihan records objective shape correspondences, not orthographic rules — having a variant does not mean it is interchangeable in your context.
A specialized variant (kSpecializedSemanticVariant) is interchangeable with the base character only under one specific sense, not across all meanings.
Variants are common in classical-text collation, name and document processing, and font development — especially the traditional/simplified issues faced in HK, Macau, Taiwan and overseas Chinese communities.
This tool runs entirely in your browser from a single locally loaded JSON file — no input is uploaded or stored, and there are no server calls.
Frequently Asked Questions
-
A variant character is one with the same meaning and pronunciation but a different written form — e.g. 群/羣, 峰/峯, 裡/裏. They are alternate shapes of the same word, not misspellings. They arise from component rearrangement, simplified-and-full forms coexisting, and centuries of copying, printing, name-taboo avoidance and regional usage.
-
Entirely from the Unicode® Unihan database (Unihan_Variants.txt), the authoritative character-property dataset maintained by the Unicode Consortium, covering tens of thousands of Han characters and used worldwide by operating systems, fonts and input methods. It is contributed jointly by the standards bodies of China, Japan, Korea, Vietnam and others. This tool only displays it locally and adds nothing.
-
Semantic variant = kSemanticVariant (interchangeable meaning); simplified/traditional = kSimplifiedVariant / kTraditionalVariant (e.g. 国↔國); Z-variant = a near-identical shape that could be unified but is encoded separately (kZVariant); specialized = interchangeable only under one sense (kSpecializedSemanticVariant); spoofing = a confusable look-alike that may be used for deception (kSpoofingVariant).
-
Within Unihan, simplified/traditional correspondences are recorded in dedicated kSimplifiedVariant and kTraditionalVariant fields and count as a kind of variant. This tool shows them under their own "Simplified form" and "Traditional form" groups — e.g. looking up 国 shows traditional 國, and looking up 変 shows its counterpart 變.
-
Not necessarily. Unihan records objective shape correspondences, not orthographic recommendations. A variant existing in the database does not mean you may freely substitute it in your specific context (a simplified-Chinese standard, textbook usage, or formal documents). Specialized variants are interchangeable only under one sense. Judge against the prevailing national/regional standard and your purpose.
-
Because the Unihan database records no variant relationship for that character — many common characters simply have no recorded variants, or their variants are not in the Unicode standard. The tool then shows "No recorded variants for this character." That does not prove it never had a historical variant; it only means the authoritative database has no matching entry right now.
-
Yes. Enter multiple characters (space-separated) or even paste a whole sentence — the tool extracts the Chinese characters, removes duplicates, and builds a variant card for each. Every variant in the results is clickable so you can keep tracing.
-
No. The tool runs entirely in your browser; the variant data is a local JSON file loaded once. Lookups make no server calls, and nothing you type is uploaded or stored.
-
Spoofing variants (kSpoofingVariant) and Z-variants are look-alikes that are hard to tell apart. Attackers can use them to craft homograph URLs or account names that trick users into thinking a fake site is genuine. Developers and security teams use this kind of data to detect and filter glyphs that could be abused for impersonation.
-
Cross-check first. The tool shows Unihan's objective shape correspondences, which make a good starting reference for research, collation and development; but publishing, textbooks and formal documents each have their own orthographic standards. Defer to the current national/regional standard — this tool is not an orthographic recommendation.
Related News
You may be interested in these recent stories from our newsroom.
-
AWS Commits $1 Billion to Embedded AI Engineers as the Enterprise Fight Shifts to Deployment
AWS has committed US$1 billion to a Forward Deployed Engineering organisation that embeds its engineers inside customer teams, following pri...
-
ByteDance Launches Seedream 5.0 Pro, an Image Model That Outputs Editable Layers
ByteDance has launched Seedream 5.0 Pro, a professional image-generation and editing model that breaks a single render into more than ten ed...
-
Mira Murati's Thinking Machines Releases Inkling, Its First Open-Weight Model
Mira Murati's Thinking Machines has released Inkling, its first model: an open-weight, multimodal Mixture-of-Experts system with 975 billion...