Model•active
Qwen3-Embedding-0.6B
Qwen3-Embedding-0.6B is a 0.6B-parameter open-weight text embedding model for multilingual retrieval, semantic search and code retrieval.
Official siteVerified Oct 4, 2026
Overview
It is the compact representative of the Qwen3 Embedding family, with a 32K sequence length and 1,024-dimensional embeddings with Matryoshka Representation Learning support.
Model specifications
- Family
- Qwen3 Embedding
- Context
- 32,000 tokens
- Input
- Text
- Output
- Embedding
- Open weights
- Yes
- Capabilities
- Embeddings