Qwen Image 2.1 Turbo
Qwen Image 2.1 Turbo is Alibaba's high-speed, cost-optimized edition of the flagship Qwen Image 2.1 vision model. Engineered on a refined 7B single-stream DiT architecture, it compresses generation down to just 8 denoising steps without sacrificing output fidelity. Qwen Image 2.1 Turbo unifies rapid 2K image rendering, native transparent background (RGBA) generation, and intuitive multi-image reference editing into a single stream.
Related Models
Qwen Image 2.1 Pro
Qwen Image 2.1 Pro is Alibaba's flagship unified model for high-fidelity text-to-image synthesis and multi-modal image editing. Built on an optimized 7B single-stream DiT architecture, it seamlessly merges studio-grade 2K image rendering, multi-image reference control (up to 10 input images), and localized inpainting via mask or brush interactions while strictly preserving subject identity.
Qwen Image 3.0 Spicy
Qwen Image 3.0 Spicy is a flagship uncensored image generation model built upon Alibaba's advanced Qwen vision architecture. Operating as a Spicy uncensored model, it completely bypasses standard content filters to unlock maximum creative freedom across photorealistic portraiture, complex visual storytelling, and high-detail stylized concept art without content restrictions. Leveraging enhanced rendering capacities, it delivers sharp textures, realistic lighting physics, and exceptional prompt adherence. Qwen Image 3.0 Spicy is the premier choice for professional designers, digital creators, marketing agencies, and visual artists requiring unconstrained, high-quality image synthesis.
Qwen Image 3.0 Pro Spicy
Qwen Image 3.0 Pro Spicy is a flagship uncensored image generation model built upon Alibaba's advanced Qwen vision architecture. Operating as a Spicy uncensored model, it completely bypasses standard content filters to unlock unrestricted artistic freedom across photorealistic portraiture, complex visual storytelling, and high-detail stylized concept art. Leveraging Pro-tier enhancements, it delivers ultra-sharp textures, master-level lighting physics, fine typography rendering inside images, and exceptional prompt adherence. Qwen Image 3.0 Pro Spicy is the premier choice for professional designers, digital creators, marketing agencies, and visual artists requiring unconstrained, studio-quality image synthesis.
Qwen Image 3.0
Qwen Image 3.0 is Alibaba's next-generation open-weights AI image generation and editing model built on a Diffusion Transformer (DiT) architecture. Supporting both text-to-image (T2I) and image-to-image instruction editing (I2I), it excels at rich content rendering, hyper-realistic detail, and deep contextual understanding. It achieves industry-leading text-in-image typography, rendering clear small text down to 10px across 12 native languages and 20+ fonts for complex multi-column layouts, UI designs, and graphic documents. With open weights available for self-hosting, Qwen Image 3.0 provides highly cost-effective, high-precision visual generation for localized advertising, product UI prototyping, e-commerce graphics, and creative publishing.
Read Me
Qwen Image 2.1 Turbo API
Qwen Image 2.1 Turbo is an image generation and editing model developed by Alibaba (Tongyi Qianwen Qwen Team). It is optimized for fast inference, high concurrency, and low latency. It uses an 8-step sampling process, compressing the standard 40 steps into 8, which is approximately 2.5 times faster than the Pro version. It excels in text-to-image (T2I), image-to-image, and instruction-based image editing.
On the iCreat platform, this model is accessed via an asynchronous two-step API using a unified task_id. The endpoint is https://api.icreat.ai/v1/task/submit/aliyun/qwen-image-2-1-turbo. It supports a resolution range from 512×512 to 2048×2048 with aspect ratios between 1:8 and 8:1. Pricing is a flat rate of $0.016 per output image for both 1K and 2K resolutions, with no additional charges for input reference images.
Model Positioning
Qwen Image 2.1 Turbo is positioned as a high-speed, cost-effective, and lightweight version within the Qwen Image series. It is designed for scenarios requiring rapid generation and high throughput, offering a unified pricing model of $0.016 per image regardless of 1K or 2K resolution. It is the ideal choice for developers needing fast iteration and scalable image generation without compromising on core capabilities like multi-image referencing and transparent backgrounds.
Core Capabilities
Text-to-Image and Image Editing
Supports text-to-image generation, image-to-image transformation, and instruction-based image editing. Users can provide text prompts and optional reference images to guide the generation process.
Native Transparent Backgrounds
Capable of natively generating images with transparent backgrounds (RGBA), eliminating the need for post-processing matting steps.
Multi-language Text Rendering
Accurately renders text within generated images, supporting multiple languages for complex typographic requirements.
Multi-Image Referencing
Accepts up to 10 reference images in a single request, enabling complex image generation and editing conditioned on multiple visual inputs.
Flexible Resolutions and Aspect Ratios
Supports total pixel dimensions between 512×512 and 2048×2048, with flexible aspect ratios ranging from 1:8 to 8:1, natively outputting up to 2K resolution.
Pricing
| Size Tier | Unit Price (USD/Image) |
|---|---|
| 1K: 10241024, 1280720, 7201280, 1024768, 768*1024 | $0.016 |
| 2K: 20482048, 25601440, 14402560, 20481536, 1536*2048 | $0.016 |
Billing Note: Billed per output image at a flat rate of $0.016 for both 1K and 2K tiers. No charge for input reference images. The official Alibaba Cloud Bailian price is ¥0.10/image, but the iCreat platform price of $0.016/image prevails.
Application Scenarios
- E-commerce product image generation and background replacement
- Marketing material and advertisement creative generation
- Rapid prototyping for design concepts
- Multi-image conditional editing for complex compositions
- Graphic design with native transparent background assets
- High-throughput batch image generation
Model Comparison
Comparison Table 1: Qwen Image Series
| Model | Capabilities | Reference Images | Transparent Output | Pricing |
|---|---|---|---|---|
| Qwen Image 2.1 Turbo (This Model) | T2I + I2I/Editing | Up to 10 | Native RGBA | 1K/2K $0.016/Image flat |
| Qwen Image 2.1 Pro | T2I + I2I/Editing | Up to 10 | Native RGBA | 1K/2K $0.04/Image flat |
| Qwen Image 3.0 Pro | T2I + I2I/Editing | 1-3 | Native RGBA | 1K $0.04, 2K $0.075/Image |
Note: Turbo is optimized for speed (8-step sampling) and cost, while Pro versions offer higher quality and detail.
Comparison Table 2: Cross-Model Comparison
| Model | Pricing | Reference Images | Output Resolution |
|---|---|---|---|
| Qwen Image 2.1 Turbo (This Model) | $0.016/Image flat | Up to 10 | 2K |
| Seedream 4.0 | Per image | Multi-image | 2K/4K |
| GPT Image 2 | Per image | Strong instruction following | Standard |
Note: Cross-model comparison focuses on capability dimensions. For exact pricing of other models, please refer to the platform page.
Why Choose Qwen Image 2.1 Turbo?
- High-speed inference with 8-step sampling, approximately 2.5x faster than the Pro version
- Flat rate pricing of $0.016 per image for both 1K and 2K resolutions
- Supports up to 10 reference images for complex conditional generation
- Native RGBA transparent background generation without post-processing
- Flexible aspect ratios from 1:8 to 8:1 and resolutions up to 2K
- Unified asynchronous API architecture for easy integration
Specifications
| Field | Value |
|---|---|
| Model Name | Qwen Image 2.1 Turbo |
| Developer | Alibaba (Tongyi Qianwen Qwen Team) |
| Model ID | aliyun/qwen-image-2-1-turbo |
| Endpoint | https://api.icreat.ai/v1/task/submit/aliyun/qwen-image-2-1-turbo |
| Task Result Endpoint | https://api.icreat.ai/v1/task/result |
| Model Type | Image Generation & Editing Model |
| Authentication | Authorization: Bearer ${ICREAT_API_KEY} |
| Input Modalities | Text prompts, Reference images |
| Reference Image Limit | 1 to 10 images |
| Output Size Range | 512×512 to 2048×2048 total pixels |
| Aspect Ratio Range | 1:8 to 8:1 |
| Output Format | Image (type: Image, contains url and download_url) |
| Billing Unit | USD per image (billed per output image; $0.016 flat for 1K and 2K) |
| API Mode | Asynchronous two-step (Submit task -> Poll result) |
Architecture
Qwen Image 2.1 Turbo is built on a unified multimodal diffusion architecture optimized for fast inference. It compresses the standard 40-step sampling process into 8 steps, significantly reducing latency while maintaining generation quality. The API uses an asynchronous two-step workflow: first submitting a task via the endpoint, then polling the result endpoint using the same task_id. The request body consists of two top-level sibling fields: input (containing the messages array with text and image content) and parameters (containing size and watermark configurations). These fields must not be nested.
Notes
- Both steps use the same task_id; while processing, the result array is empty, and you should continue polling.
- FAILED is a terminal state; check your request parameters or reference image resources.
inputandparametersare top-level sibling fields and must not be nested within each other.- Supports 1 to 10 reference images per request.
sizedefaults to "1024*1024"; total pixels must be between 512×512 and 2048×2048, with aspect ratios between 1:8 and 8:1.watermarkdefaults to false.costUSDis 0.016 (USD) per successful generation and is only returned upon success.- 1K and 2K tiers have the same billing of $0.016/image; input reference images are not charged extra.
Frequently Asked Questions
How do I call Qwen Image 2.1 Turbo?
It uses a two-step asynchronous process. First, submit the task to the endpoint https://api.icreat.ai/v1/task/submit/aliyun/qwen-image-2-1-turbo using Bearer authentication. Then, poll the result endpoint https://api.icreat.ai/v1/task/result with the returned task_id until the status is SUCCEEDED or FAILED.
How is billing calculated?
Billing is based on the number of output images generated, at a flat rate of $0.016 per image for both 1K and 2K resolutions. Input reference images are not charged. For example, generating 2 images at 2048*2048 will cost $0.032. The costUSD field in the response will indicate the cost of the successful generation.
How many reference images are supported?
It supports 1 to 10 reference images, which can be used for image-to-image generation, instruction-based editing, or multi-image conditional generation.
What sizes are supported?
The size parameter defaults to "1024*1024". The total pixel dimensions must be between 512×512 and 2048×2048, and the aspect ratio must be between 1:8 and 8:1.
Does it support transparent backgrounds?
Yes, it supports native transparent background generation in RGBA format.
What is the difference between Qwen Image 2.1 Turbo and Qwen Image 2.1 Pro?
Qwen Image 2.1 Turbo features a flat rate of $0.016 per image, focusing on speed and high concurrency with 8-step sampling. Qwen Image 2.1 Pro has a flat rate of $0.04 per image, offering higher quality and detail. Both support up to 10 reference images and native RGBA transparent outputs.