Qwen3-VL-Embedding is a unified multimodal vector model based on Qwen3-VL, supporting text, image, video (single or mixed modality) input. It outputs unified representation vectors, suitable for cross-modal retrieval, image/video search, clustering, complex multimodal retrieval, and tagging.
Inference Service Provider
The inference service provider for qwen3-vl-embedding is QwenCloud.
Model Capabilities
| Capability | Support | Capability | Support |
|---|---|---|---|
Input Modality | Text Image | Output Modality | — |
Model Experience | Supported | Function Calling | Unsupported |
Structured Outputs | Unsupported | Web Search | Unsupported |
Prefix Completion | Unsupported | Context Caching | Unsupported |
Batch Inference | Unsupported | Fine-tuning | Unsupported |
Context Limits
| Parameter | Value | Parameter | Value |
|---|---|---|---|
Max Input Length | 32000 | Max Output Length | — |
Context Window | — |
Pricing
This page only shows the original pricing for model API calls, excluding any limited-time promotions. Visit QwenCloud console for promotional offers.
- China North 2 (Beijing)
| Billing Item | Price (USD) | Unit |
|---|---|---|
Image Input | 0.258 | Per 1M tokens |
Text Input | 0.1 | Per 1M tokens |
Rate Limits
- China North 2 (Beijing)
| Parameter | Value |
|---|---|
RPM (Requests Per Minute) | 2400 |
TPM (Tokens Per Minute) | 1,200,000 |