Qwen3 EmbeddingUpdate
Qwen3 Embedding and Reranker
The update, in brief
Qwen3 releases embedding and reranking models for multilingual retrieval.
What’s new
- 0.6B, 4B and 8B variants use Apache 2.0.
Availability details
These models return vectors or relevance scores, rather than chat answers.
Availability and conditions refer to the original announcement.
Original sources
1Read the announcement, watch the demo, or hear from the team.
More about this release
Demos, launch pages and discussions explicitly connected to this announcement.
- Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models
Original page reviewed
Open originalLink checked - Qwen3-Embedding-8B · Hugging Face
Original page reviewed
Open originalLink checked
Summarized from the linked sources. Availability refers to the announcement date.
Report a correction
Tell us what needs correcting. Reports are reviewed before any public content changes.