Lightweight 0.6B multilingual embedder with 32K context and instruction support.
Copy and paste this command to start running the model locally.
View on Hugging Face
Run with Ollamaollama run qwen3-embedding:0.6bAccess model weights, configuration files, and documentation.
See which devices can run this model and at what quality level.
The smallest member of the Qwen3-Embedding series, fine-tuned from Qwen3-0.6B-Base via the same 3-stage contrastive + supervised + model-merging pipeline as its larger siblings. Supports 100+ languages, 32K context, instruction-aware queries and Matryoshka output dimensions, making it a popular choice for edge and resource-constrained deployments.