Models/Qwen Image/Image API

Live15 endpoints

Qwen Image 2 API — Qwen Image 2.0 Text-to-Image & Editing

The Qwen Image 2 API gives you programmatic access to Alibaba's Qwen Image 2.0 model family — text-to-image, instruction-based editing, a higher-fidelity Pro tier, plus the original Qwen Image line and LoRA training/support endpoints, all through one Muapi API key, from $0.02 per request.

Get Qwen Image API Key Try Qwen Image 2.0 Pro in Playground

🆕 Qwen Image 2.0The latest Qwen Image generation, with a higher-fidelity Pro tier4 models

Text to ImageNew

🆕 Qwen Image 2.0

Qwen Image 2.0

Qwen 2.0 Text-to-Image sets a new standard for AI-driven visual creation, combining complex prompt understanding with high-fidelity, photorealistic output.

Image

$0.04/request

Try Model

Image to ImageNew

🆕 Qwen Image 2.0

Qwen Image 2.0 Edit

Qwen 2.0 Edit provides detailed, natural modifications of images through simple text prompts.

Image

$0.04/request

Try Model

Text to ImageNew

🆕 Qwen Image 2.0

Qwen Image 2.0 Pro

Qwen 2.0 Pro Text-to-Image offers maximum fidelity and instruction-following for professional-grade creations.

Image

$0.09/request

Try Model

Image to ImageNew

🆕 Qwen Image 2.0

Qwen Image 2.0 Pro Edit

Qwen 2.0 Pro Edit allows extreme, precise edits while preserving original photorealism.

Image

$0.09/request

Try Model

🖼️ Qwen Image (Original)The original Qwen Image text-to-image and editing models4 models

Text to ImageClassic

🖼️ Qwen Image (Original)

Qwen Image

The original Qwen Image text-to-image model, converting detailed textual prompts into high-quality images.

ImageFast

$0.03/request

Try Model

Image to ImageClassic

🖼️ Qwen Image (Original)

Qwen Image Edit

Transform an existing image using a natural-language edit prompt with the original Qwen Image editing model.

ImageFast

$0.03/request

Try Model

Image to ImageClassic

🖼️ Qwen Image (Original)

Qwen Image Edit Plus

An expanded Qwen Image editing endpoint that accepts multiple reference images plus explicit width/height control.

ImageFast

$0.03/request

Try Model

Image to ImageClassic

🖼️ Qwen Image (Original)

Qwen Image Edit 2511

A refreshed Qwen Image editing checkpoint with multi-image reference support and explicit width/height control.

ImageFast

$0.04/request

Try Model

🎛️ LoRA SupportGeneration and editing endpoints that accept a custom LoRA5 models

LoRA SupportNew

🎛️ LoRA Support

Qwen Image Text-to-Image (LoRA)

Generate an image from a text prompt while applying one or more custom-trained Qwen-Image LoRAs.

ImageFast

$0.02/request

Try Model

LoRA SupportNew

🎛️ LoRA Support

Qwen Image Text-to-Image 2512 (LoRA)

Generate an image on the 2512 Qwen-Image checkpoint while applying one or more custom-trained LoRAs.

ImageFast

$0.02/request

Try Model

LoRA SupportNew

🎛️ LoRA Support

Qwen Image Edit Plus (LoRA)

Edit an image from multiple reference images while applying one or more custom-trained Qwen-Image LoRAs.

ImageFast

$0.04/request

Try Model

LoRA SupportNew

🎛️ LoRA Support

Qwen Image Edit 2511 (LoRA)

Edit an image on the 2511 checkpoint while applying one or more custom-trained Qwen-Image LoRAs.

ImageFast

$0.04/request

Try Model

LoRA SupportNew

🎛️ LoRA Support

Qwen Image Edit (LoRA)

Edit an image with the original Qwen Image editing model while applying a custom-trained LoRA.

ImageFast

$0.04/request

Try Model

🏋️ LoRA TrainerFine-tune a custom Qwen-Image LoRA from your own dataset2 models

TrainingNew

🏋️ LoRA Trainer

Qwen-Image LoRA Trainer

Accelerated fine-tuning for Qwen-Image models directly from a custom dataset zip archive — train character, style, or concept LoRAs.

LoRA weights

$2.00/request

Try Model

TrainingNew

🏋️ LoRA Trainer

Qwen-Image-2512 LoRA Trainer

Fine-tune a custom LoRA on the 2512 Qwen-Image checkpoint from your own dataset zip archive.

LoRA weights

$2.00/request

Try Model

What is the Qwen Image 2 API?

The Qwen Image 2 API gives applications programmatic access to Alibaba's Qwen Image 2.0 model — text-to-image generation and instruction-based editing with strong prompt understanding and photorealistic output, plus a higher-fidelity Pro tier for professional-grade work.

Muapi exposes the full Qwen Image family — 2.0, the original Qwen Image line, and LoRA training/support endpoints — through one REST API. Use the image playground to compare Qwen Image 2.0 against the original line, then train and apply a custom LoRA for consistent characters or styles in your own pipeline.

Key Capabilities

Text-to-Image

Turn a text prompt into a high-fidelity, photorealistic image with strong prompt-following.

Instruction-Based Editing

Modify an existing image with a natural-language edit prompt instead of masks or layers.

Pro Tier for Professional Work

Step up to Qwen Image 2.0 Pro for maximum fidelity and instruction-following on professional-grade creations.

Multi-Image Reference Editing

Qwen Image Edit Plus and Edit 2511 accept multiple reference images with explicit width/height control.

Custom LoRA Training

Fine-tune a Qwen-Image LoRA from your own dataset zip archive for consistent characters, styles, or concepts.

LoRA-Aware Generation & Editing

Apply one or more custom-trained LoRAs directly on generation and editing endpoints, not just at training time.

Qwen Image API Endpoint Comparison

Endpoint	Input	Output	Price	Best For
qwen-image-2.0	Prompt	Image	$0.04/request	Standard-quality generation from text
qwen-image-2.0-edit	Image + prompt	Image	$0.04/request	Standard-quality instruction editing
qwen-image-2.0-pro	Prompt	Image	$0.09/request	Maximum fidelity from text
qwen-image-2.0-pro-edit	Image + prompt	Image	$0.09/request	Precise edits preserving photorealism
qwen-image	Prompt	Image	$0.03/request	Existing Qwen Image (original) workflows
qwen-image-edit	Image + prompt	Image	$0.03/request	Existing Qwen Image (original) edit workflows
qwen-image-edit-plus	1+ images + prompt	Image	$0.03/request	Multi-reference edits with size control
qwen-image-edit-2511	1+ images + prompt	Image	$0.04/request	Refreshed checkpoint multi-reference edits
qwen-image-text-to-image-lora	Prompt + LoRA	Image	$0.02/request	Custom character/style generation
qwen-image-edit-lora	Image + prompt + LoRA	Image	$0.04/request	Custom character/style editing
qwen-image-lora-trainer	Dataset zip	LoRA weights	$2.00/request	Training a new custom LoRA

Prices and supported fields are model-specific. Review the live Playground schema before shipping generated images in production.

How to Call the Qwen Image API

Submit a JSON request with your Muapi API key, save the request ID, and poll the shared result endpoint. Provide a publicly reachable image URL for edit requests.

1. Generate an image from text

curl -X POST https://api.muapi.ai/api/v1/qwen-image-2.0-pro \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"A minimalist product shot of a ceramic mug on a marble countertop, soft studio lighting","aspect_ratio":"1:1"}'

# Response: {"request_id":"REQUEST_ID"}

Call qwen-image-2.0-pro with a prompt and aspect ratio.

2. Edit an image

curl -X POST https://api.muapi.ai/api/v1/qwen-image-2.0-pro-edit \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"Change the background to a sunlit kitchen","images_list":["https://example.com/reference.jpg"],"aspect_ratio":"1:1"}'

# Response: {"request_id":"REQUEST_ID"}

Call qwen-image-2.0-pro-edit with a reference image URL and an edit prompt.

3. Poll the result

curl https://api.muapi.ai/api/v1/predictions/REQUEST_ID/result \
  -H "x-api-key: YOUR_API_KEY"

# Poll until status is completed, then read the generated image URL from outputs.

Poll until the task is completed, then read the returned image URL.

Frequently Asked Questions

Can I generate an image from a text prompt with the Qwen Image API?

Yes. Both qwen-image and qwen-image-2.0 accept a text prompt and return a generated image; qwen-image-2.0-pro adds maximum fidelity and instruction-following for professional work.

What's the difference between Qwen Image and Qwen Image 2.0?

Qwen Image 2.0 is the newer generation, with stronger complex-prompt understanding and higher-fidelity photorealistic output. The original Qwen Image line remains available for existing workflows.

What's the difference between the Standard and Pro tiers?

Standard-tier endpoints (qwen-image-2.0, qwen-image-2.0-edit) cost $0.04 per request; the Pro tier (qwen-image-2.0-pro, qwen-image-2.0-pro-edit) costs $0.09 per request and offers maximum fidelity and instruction-following.

Can I edit an image with multiple reference images?

Yes. qwen-image-edit-plus and qwen-image-edit-2511 accept a list of reference images plus explicit width and height control, instead of a single image URL.

Can I train a custom LoRA on Qwen Image?

Yes. qwen-image-lora-trainer and qwen-image-2512-lora-trainer fine-tune a custom LoRA from a dataset zip archive, for consistent characters, styles, or concepts.

Can I use a custom LoRA when generating or editing, not just at training time?

Yes. The LoRA-support endpoints (qwen-image-text-to-image-lora, qwen-image-edit-lora, and their variants) accept one or more custom-trained LoRAs directly on generation and editing requests.

How much does the Qwen Image API cost?

LoRA-support generation/editing costs $0.02–$0.04 per request; the original Qwen Image line costs $0.03–$0.04; Qwen Image 2.0 costs $0.04 (Standard) or $0.09 (Pro) per request; LoRA training costs $2.00 per request.

Compare AI Image APIs →See Qwen Image alongside Nano Banana, Seedream, and every other image model on Muapi.Explore the AI LoRA Trainer API →Train custom LoRAs across Qwen-Image, Z-Image, and other supported base models.Browse text-to-image models →Compare every text-to-image model in the Muapi catalog side by side.

Ready to build with Qwen Image?

Create one Muapi API key and connect text-to-image, instruction editing, multi-reference editing, and custom LoRA training in a single asynchronous workflow.

Get an API Key Try Qwen Image 2.0 Pro in Playground