Qwen3 Embedding

An open source model from the Qwen family for generating embeddings — vectors that represent the meaning of a text, the foundation for semantic search, deduplication, clustering, and RAG.

id	qwen3-embedding
Type	Embeddings
Max input	8K tokens
Available on	Starter, Standard, Pro

When to use

• Semantic search over your documents
• RAG (retrieval-augmented generation)
• Group or deduplicate texts by similarity

Example

Python

resp = client.embeddings.create(
    model="qwen3-embedding",
    input=["how to reset my password", "forgot the password"],
)
print(resp.data[0].embedding[:5])

curl

curl https://api.norlen.io/v1/embeddings \
  -H "Authorization: Bearer $NORLEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"qwen3-embedding","input":"how to reset my password"}'

Endpoint details in Embeddings.