text-embedding-v4

Models/text-embedding-v4

Model Code

text-embedding-v4

Model overview

Embedding model

Text Embedding V4 is a general multilingual embedding model trained by Tongyi Lab based on Qwen3. Compared with V3, it improves retrieval, clustering, and classification performance, performs better on multilingual, Chinese-English, and code retrieval benchmarks, and supports custom dimensions from 64 to 2048.

Company: AlibabaProvider: Alibaba Cloud

Model capabilities

Input modalityT

Output modalityTV

Model experience✓

Function calling-

Structured output-

Web search-

Prefix completion-

Context cache-

Batch inference✓

Fine-tuning-

Model pricing

Embedding input (Batch File)$0.03731/1M tokens

Text input$0.07463/1M tokens

Rate limits and context

Maximum input length-

Context length-

Maximum output length-

RPM1800

TPM1200000

API reference

EndpointPOST /v1/embeddings

SDKOpenAI-compatible

API code examples

Get API Key

import os
import requests

response = requests.post(
    "https://api.wlrouter.com/v1/embeddings",
    headers={
        "Authorization": f"Bearer {os.environ['WLROUTER_API_KEY']}",
        "Content-Type": "application/json",
    },
    json={
  "model": "text-embedding-v4",
  "input": "Embed this short sentence."
}
)
response.raise_for_status()
print(response.json())