The Guild Wars 2 API¶
ArenaNet provides an official read-only API for Guild Wars 2 that can be used to query information about the game. There are multiple versions of the API that are currently supported:
- Version 1 which is the originally available version, and
- Version 2 which is the currently maintained version.
The concepts explained in this document are consistent between both versions of the API. Visit the documentation for the individual versions for version-specific information.
Localization¶
The Guild Wars 2 API offers support for localized responses. The supported languages differ between API versions. (See the table below for a full comparison.)
| Language | API v1 | API v2 |
|---|---|---|
| Chinese (zh) | - | ✔ |
| English (en) | ✔ | ✔ |
| French (fr) | ✔ | ✔ |
| German (de) | ✔ | ✔ |
| Spanish (es) | ✔ | ✔ |
A specific language can be requested by setting the lang query parameter to
the language code.
$ curl https://api.guildwars2.com/v2/items?id=43766&lang=en
{
"name": "Tome of Knowledge",
"description": "Grants one character level if below level 80 or one spirit shard if level 80.\nOnly usable in PvE and WvW.",
"type": "Consumable",
"level": 0,
"rarity": "Exotic"
// additional fields were excluded from this example
}
$ curl https://api.guildwars2.com/v2/items?id=43766&lang=de
{
"name": "Foliant des Wissens",
"description": "Lässt Euren Charakter eine Stufe aufsteigen, wenn er Stufe 80 noch nicht erreicht hat, oder gewährt eine Geister-Scherbe, wenn er auf Stufe 80 ist.\nNur im PvE und WvW verwendbar.",
"type": "Consumable",
"level": 0,
"rarity": "Exotic"
// additional fields were excluded from this example
}
Note
The API makes some effort to guess the language based on your IP geolocation. Hence, it is recommended to always request a language explicitly for consistent results.
Rate Limiting¶
The Guild Wars 2 API implements basic rate limiting. The rate limit is dependent on the request's IP address and applies to all endpoints whether the endpoint is authenticated or not.
The limit is implemented using a refilling token bucket. The bucket can hold a maximum of 300 tokens and is fully replenished after one minute of being depleted.
When the limit is reached, subsequent requests will yield a status 429 response.