Qwen3-Embedding-4B
Qwen/Qwen3-Embedding-4B
A popular open embeddings model, with 2.6M downloads a month. gigarouter benchmarks and hosts it as an OpenAI-compatible API.
about this model
Qwen3-Embedding-4B is a 4-billion-parameter text embedding model from the Qwen3 Embedding series, optimized for multilingual text embedding and ranking tasks. It is designed for text retrieval, code retrieval, classification, clustering, and bitext mining across more than 100 languages.
Key Capabilities
- Context length: 32K tokens.
- Embedding dimension: Up to 2,560; supports Matryoshka Representation Learning (MRL) for user-defined dimensions from 32 to 2,560.
- Instruction-aware: Both embedding and reranking models accept task-specific instructions; using instructions typically improves downstream task performance by 1%–5%.
- Multilingual: Supports 100+ natural languages and multiple programming languages for code retrieval.
Performance
The 8B variant of the Qwen3 Embedding series achieved No. 1 rank on the MTEB multilingual leaderboard as of June 5, 2025, with a score of 70.58. The Qwen3-Embedding-4B benefits from the same model architecture and training methodology, providing strong multilingual and cross-lingual retrieval quality.
Model Series Overview
| Model Type | Models | Size | Layers | Sequence Length | Embedding Dimension | MRL Support | Instruction Aware |
|---|---|---|---|---|---|---|---|
| Text Embedding | Qwen3-Embedding-0.6B | 0.6B | 28 | 32K | 1024 | Yes | Yes |
| Text Embedding | Qwen3-Embedding-4B | 4B | 36 | 32K | 2560 | Yes | Yes |
| Text Embedding | Qwen3-Embedding-8B | 8B | 36 | 32K | 4096 | Yes | Yes |
| Text Reranking | Qwen3-Reranker-0.6B | 0.6B | 28 | 32K | – | – | Yes |
| Text Reranking | Qwen3-Reranker-4B | 4B | 36 | 32K | – | – | Yes |
| Text Reranking | Qwen3-Reranker-8B | 8B | 36 | 32K | – | – | Yes |
For further details, refer to the official blog post and GitHub repository.
We're benchmarking and onboarding Qwen3-Embedding-4B as a hosted, OpenAI-compatible API. Sign in for free credit and be ready when it lands, or tell us you want it and we'll prioritize it.