Qwen3-Embedding-4B is the latest proprietary model in the Qwen3 family of embedding models, designed specifically for text embedding and sorting tasks. The model is based on the Qwen3 series of dense base models with a 4 billion parameter scale, supports context lengths of up to 32K, and can generate embedding vectors of up to 2560 dimensions. The model inherits the excellent multi-language capabilities of the basic model, supports more than 100 languages, and has long text...
Provider: Qwen
Context window: 32K
Pricing: $0.02 input / $0.00 output per 1M tokens
Capabilities: serverless
