What is MergeRef
MergeRef is a desktop workspace for game and anime artists. It puts your references, your AI models and your finished shots on one infinite canvas, so the whole pipeline from reference gathering to final render happens in one place.
Who it is for
MergeRef is built for character artists, concept artists, illustrators and small studios who already work with reference boards and want AI to handle the repetitive parts: rough coloring, outfit and pose swaps, lighting passes, design variations and short motion tests. You stay in control of every input; the models do the mechanical work.
The workspace at a glance
| Area | What it does |
|---|---|
| Infinite canvas | The board where every image, video and result lives. Pan and zoom without limits. Learn more |
| Shared toolbar | One toolbar for canvas tools and AI tools. The active tool decides which options appear. Learn more |
| Selection bar | Floats above a selected image. Assigns roles such as ORIG, REF, costume, pose and style. Learn more |
| Prompt panel | Where you write the prompt, pick a model, set options and press Generate. Learn more |
| History panel | Every generation, with the exact prompt, model and inputs used. Learn more |
| Source image panel | Shows which images will be sent with the next generation, grouped by role. Learn more |
How generation works
MergeRef runs on cloud models from Google (Gemini, Veo), Black Forest Labs (FLUX), OpenAI (GPT Image) and ByteDance (Seedream, Seedance). You paste your own API key and pay the provider directly. MergeRef adds no markup.
Where to go next
- New to MergeRef: follow the Quickstart to make your first generation in about five minutes.
- Choosing a model: read the Model overview for strengths and costs side by side.
- Setting up billing: see API keys & billing.