Images from a description
Generate concepts from a prompt. Define scene, composition and style to compare images against your project brief.
Use cases / Image and video
Image generation
Generate and edit AI images inside your own product.
Turn a description into an image, transform existing content and create variations of your ideas. Integrate AI image generation into creative tools, products and content workflows.
With QDivZero, you can deploy compatible text-to-image and instruction-based editing models. Choose a model, connect your application through an API and use the output without setting up and maintaining GPUs yourself.
From an idea to an image
Generate the content you need, when you need it.
Create, edit and explore visual concepts within your product. Choose a variant compatible with each task’s inputs.
Editing through instructions
Modify an image with a compatible editing model. Explain what should change and which elements you want to preserve.
Reference variations
Explore alternatives with models that accept references. Compare composition and style while retaining your project’s context.
Generation through an API
Connect generation to your application through an API. Organise requests, statuses and versions to continue the creative workflow.
Build it with QDivZero
From a prompt to an image for your application.
Choose a variant supporting creation, editing or references. Send instructions and inputs by API; your application organises jobs, versions and results to continue the creative workflow.
Your application flow
Prompt and references
Send text and images supported by the model.
Configuration
Set size, style, and available generation parameters.
Generation
The model creates or edits the requested image.
Image
Receive the result and use it in your application.
Open-weight / Hugging Face
You can start with…
Compare text-to-image and editing models with your prompts and references. Quality, controls and available formats depend on the variant; choose based on visual results and the capacity your application needs. Controls vary between model variants. Compare generation, editing, and reference use as different tasks and choose according to the outputs your tool should let people create.
Image generation
Z-Image-Turbo
Explore image generation with a model designed for fast inference.
View model on Hugging FaceImage editing
Qwen-Image-Edit
Modify existing images using text instructions.
View model on Hugging FaceImage generation
Image generation: frequently asked questions
How do I generate AI images from text?
Use a text-to-image model and send a prompt describing content, style, and composition. Deploy compatible models on QDivZero and connect them to your application. Parameters and output formats depend on your chosen generation model.
Which open-weight models can generate images?
Start by comparing Qwen-Image and Z-Image-Turbo for generation from text. Check each version model card and requirements for a compatible deployment. Choose based on visual quality, available controls, and application capacity needs.
Can I edit images with AI text instructions?
Yes. An editing model such as Qwen-Image-Edit receives an image and a description of the change. Integrate this task into your product to modify visual content. Transformations and controls depend on the model version.
How do I integrate an image generation API into my product?
Deploy a compatible model on QDivZero and send application prompts to its endpoint. Configure supported parameters and use the returned image in your product. Your application organizes requests and output storage.
Do I need to manage GPUs to generate images with my own models?
QDivZero prepares the environment and compute for a compatible model. Choose Hugging Face or private and fine-tuned models without configuring GPUs directly. Check model requirements and available deployment capacity.
What determines AI image generation cost?
Cost depends on the model, resolution, generation parameters, and compute used. Image volume and concurrency also influence capacity needs. Compare configurations with your prompts and review QDivZero pricing options.
Can I generate images from a reference image?
Yes, if you use a variant that accepts references or image inputs. Specify the features you want to preserve and the changes you need. Check whether the model supports editing, variations or another task, since each uses different inputs and controls.
How do I choose an image generation model for my product?
Compare your usual scenes and styles, text quality within images and instruction following. Evaluate resolution, generation time and support for editing or references. Check the license of your chosen variant for the intended product use.
Ready to create images from your application?
Integrate image generation and editing with the models you choose. QDivZero prepares the environment and compute for your compatible deployment so you can connect prompts, references and results with your product.