text-embedding-v4

Active

qwen/text-embedding-v4

是通义实验室基于Qwen3训练的多语言文本统一向量模型，相较V3版本在文本检索、聚类、分类性能大幅提升；在MTEB多语言、中英、Code检索等评测任务上效果提升15%~40%；支持64~2048维用户自定义向量维度。

EmbeddingQwen

Pricing

Pay as you go · priced in CNY

Pay as you go · priced in CNYInput tokens

Input

¥0.500

/ 1M tokens

Embedding models are billed on input tokens only.

Providers and quotes

1 routes · pay as you go

Provider	Input	Context length	Latency	Uptime	Status
dashscope dashscope/text-embedding-v4	¥0.500	—	460ms	0.60%

API examples

OpenAI compatible

Endpoint/embeddingsmodel: qwen/text-embedding-v4

python

from openai import OpenAI

client = OpenAI(
  base_url="https://api.openroute.cn/v1",
  api_key="<ONTOROUTE_API_KEY>",
)

response = client.embeddings.create(
  model="qwen/text-embedding-v4",
  input="The food was delicious and the waiter was friendly.",
)

print(response.data[0].embedding)

OpenRoute exposes an OpenAI-compatible Embeddings API. Call it directly or via the OpenAI SDK.

Request parameters

Calls (30d)

1 supply routes

Avg. latency

460ms

Time to first token

Usage trend

近 14 天 8 tokens

10-0910-22

Connectivity

Last 84 hours · Proxy probes