AlibabaQwen Embedding
Qwen3-Embedding-8B
This instruction-aware embedding checkpoint supports more than 100 languages and a 32,000-token context.
Practical use
Practical guidance. A wider vector provides more storage choices, but the maximum width is not automatically best for a given collection. Measure relevant-result recall and index size together using real queries.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | Downloadable |
| Model identifiers | Qwen/Qwen3-Embedding-8B |
| Vector dimensions | User-selected dimensions from 32 to 4096.Qwen3-Embedding-8B model card |
| Scope | Text, cross-language and code retrieval.Qwen3-Embedding-8B model card |
| Licence | Apache 2.0.Qwen3-Embedding-8B model card |
Related models
- Qwen3-Embedding-4B
Alibaba · Qwen Embedding
Four-billion-parameter text embedding model for instruction-aware retrieval across more than 100 languages.
Launch date not confirmedDownloadable - Qwen3-Embedding-0.6B
Alibaba · Qwen Embedding
A 600-million-parameter text embedding model with task instructions, over 100 languages and a 32,000-token context.
Launch date not confirmedDownloadable - Qwen3.7 Text Embedding
Alibaba · Qwen Embedding
Hosted text embedding family with standard and Flash variants and inputs up to 131,072 tokens.
Launch date not confirmedAPI