qwen3-vl-embedding

AlibabaToken-based

千问多模态向量模型，支持文本、图像和视频输入，用于跨模态检索与相似度计算。

textimagevideoembeddingscontext:32000

API access

RequestPOST/v1/chat/completions

Example request

Parameters

Parameter	Type	Default / range	Description
`input`required	string	—	Text or array of texts to embed
`dimensions`	integer	>= 1	Truncate embeddings to this many dimensions
`encoding_format`	enum	=`float`	Wire encoding for the embedding vectors
`user`	string	—	End-user identifier for abuse monitoring

Replace <YOUR_API_KEY> with the API key from your token settings.

All requests must include Authorization: Bearer <TOKEN> header. Anthropic-formatted endpoints accept the x-api-key header instead.

Generate tokens from the Tokens page; you can scope them to specific models, groups, IPs, and rate-limits.

Generation parameters

Parameter	Type	Default / range	Description
`input`required	string	—	Text or array of texts to embed
`dimensions`	integer	>= 1	Truncate embeddings to this many dimensions
`encoding_format`	enum	=`float`	Wire encoding for the embedding vectors
`user`	string	—	End-user identifier for abuse monitoring

Supplier	RPM	TPM	RPD
Alibaba	7.8K	1.6M	157K
official	9.7K	1.9M	195K

RPM = requests per minute, TPM = tokens per minute, RPD = requests per day. Limits apply per token group.