← All models

Qwen3.7-Text-Embedding

Published Publisher Alibabaqwen3.7-text-embedding

Overview

Qwen3.7-Text-Embedding is a multilingual text embedding model based on Qwen3.7. It converts text into numerical vectors for semantic search, retrieval, recommendations, clustering and classification. Alibaba reports improved retrieval, clustering and classification performance over text-embedding-v4, including a 20% improvement on the cited MTEB multilingual, Chinese-English and code-retrieval tasks. (alibabacloud.com)

Capabilities and limits

  • Supports 201 languages and dialects. The documented output dimensions are 256, 512, 768, 1,024, 1,536, 2,048 and 2,560; the default is 1,024. (alibabacloud.com)
  • Supports dense and sparse embeddings; sparse outputs are available through the DashScope API. (alibabacloud.com)
  • The model page lists a 131,072-token context window. For the synchronous API, the documented limit is 128,000 tokens per text item and up to 20 items per request. (alibabacloud.com)
  • This is a text embedding model, not a conversational text-generation model. Function calling, web search, context caching and fine-tuning are not supported. (alibabacloud.com)

Pricing

Prices below are official API list prices checked on November 16, 2026. They apply to the stated platform and region; they are not currency conversions of the same package.

Release and sources

The official Singapore-region release announcement lists July 31, 2026, with International deployment scope. Official release announcement. (alibabacloud.com)