Qwen
Qwen: Qwen3 Embedding 4B
qwen/qwen3-embedding-4b
The Qwen3 Embedding model series is the latest proprietary model of the Qwen family, specifically designed for text embedding and ranking tasks. This series inherits the exceptional multilingual capabilities, long-text understanding, and reasoning skills of its foundational model. The Qwen3 Embedding series represents significant advancements in multiple text embedding and ranking tasks, including text retrieval, code retrieval, text classification, text clustering, and bitext mining.
/ 1M tokens · Token reference
/ 1M tokens · Token reference
tokens
Embeddings output · Token reference fields do not establish an output-unit price.
Limits
Context
32,768
tokens
Max output
29,491
90% of the context window
Parameters
4,021,774,336
Model facts
Author
Qwen
Tokenizer
Other
Modality
Text -> Embeddings
License
Apache-2.0
Created
Oct 28, 2025
Hugging Face
Qwen/Qwen3-Embedding-4B