> For the complete documentation index, see [llms.txt](https://docs.qolaba.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.qolaba.ai/image-generation/understanding-credits-usage.md). # Understanding Credits Usage Every image generation and editing action in Qolaba consumes credits. The total cost varies based on the model selected, quality setting, number of generations, and which editing tools are used. Understanding how credits are calculated helps you generate efficiently and get the most value from your credit balance. > **Page description:** How credits are calculated in Qolaba's Image Generation workspace — what affects cost, how to estimate before generating, and how to optimize credit usage across your workflow. *** #### What Affects Credit Cost **Model** Each model has a fixed credit cost per image. This is the single biggest variable in your generation cost — premium models like Nano Banana Pro cost significantly more per image than cost-efficient models like ImageGen Fast. | Model | Credits / Image | | ------------------- | --------------- | | **Nano Banana Pro** | 49 | | **Nano Banana 2** | 25 | | **GPT Image 2** | 18 | | **Flux 1.1 Pro** | 16 | | **Seedream 4.5** | 21 | | **Recraft V3** | 22 | | **Ideogram V3** | 22 | | **SD 3.5** | 23 | | **DALL-E 3** | 12 | | **SDXL** | 18 | | **Flux Dev** | 9 | | **SD 3.5 Turbo** | 8 | | **SD 3.5 Medium** | 9 | | **ImageGen 4** | 4 | | **ImageGen Fast** | 3 | **Quality** Higher quality settings increase the credit cost per image. Generating at 4K costs more than generating the same image at 720p. | Quality | Relative Cost | | -------- | ------------- | | **720p** | Lowest | | **1K** | Low–Moderate | | **2K** | Moderate–High | | **4K** | Highest | **Number of Generations** Generation count multiplies the per-image cost linearly. Generating 4 images costs exactly 4 times the cost of generating 1. **Dimensions** Dimensions do not affect credit cost. You can change aspect ratio and pixel dimensions freely without impacting your generation budget. **Editing Tools** Each image editing tool consumes credits separately from generation. Credit cost per edit is displayed before you apply the tool. | Tool | Credit Behaviour | | ---------------------- | --------------------------------------------------------- | | **Inpainting** | Credits consumed per edit based on model and area | | **Upscaling** | Credits consumed based on upscale multiplier (2x, 3x, 4x) | | **Background Removal** | Fixed credit cost per image processed | | **Image Variation** | Credits consumed per variation generated | *** #### How Total Credits Are Calculated All per-image factors combine into a single per-image cost, which is then multiplied by your generation count: > **Total credits = credits per image × number of generations** The credits per image figure reflects your combined choice of model and quality setting. The total cost for your current configuration is displayed on the **Generate button** before you confirm — review it before every run. **Example:** | Setting | Value | | --------------------- | --------------- | | Model | Nano Banana Pro | | Quality | 2K | | Number of generations | 3 | | Credits per image | 49 | | **Total credits** | **147** | Swap Nano Banana Pro for ImageGen Fast at the same settings: | Setting | Value | | --------------------- | ------------- | | Model | ImageGen Fast | | Quality | 2K | | Number of generations | 3 | | Credits per image | 3 | | **Total credits** | **9** | Model selection has by far the greatest impact on total cost per run. *** #### Reviewing Cost Before You Generate The total credit cost for your current configuration is always displayed on the **Generate button** before you confirm. It updates immediately as you change any setting — model, quality, or generation count. Make it a habit to review this figure before every run — particularly when: * Switching to a higher-cost model * Increasing generation count for a batch run * Upgrading quality to 2K or 4K for the first time with a new prompt *** #### Optimizing Credit Usage **Use Cost-Efficient Models for Testing** Reserve premium models like Nano Banana Pro for final production output. Use ImageGen Fast, Flux Dev, or SD 3.5 Turbo for prompt testing, composition validation, and style exploration — then switch to your preferred premium model once the direction is confirmed. **Example workflow:** * Test prompt and composition → **ImageGen Fast** at 720p, 1 generation * Refine keywords and style → **Flux Dev** at 1K, 1 generation * Validate final direction → **Nano Banana 2** at 1K, 1 generation * Final production batch → **Nano Banana Pro** at 2K or 4K, 2–4 generations *** **Draft vs. Final Quality Strategy** Never generate at 4K during testing. All meaningful output qualities — composition, subject accuracy, color, style — are fully visible at 720p or 1K. **Stage 1 — Draft:** Generate at 720p or 1K with a cost-efficient model. Iterate until the prompt, keywords, and overall direction are confirmed. **Stage 2 — Final:** Switch to your target model and quality setting for the confirmed output. Generate the final batch only when the direction is locked. *** **Use Upscaling Instead of Regenerating at Higher Quality** If a 1K generation looks correct compositionally but needs higher resolution for delivery, use Image Upscaling → to increase resolution without regenerating. Upscaling a confirmed 1K image to 4x is significantly more credit-efficient than regenerating at 4K from scratch. *** **Additional Credit-Saving Practices** * **Use negative keywords** to reduce failed generations — fewer unwanted outputs means less regeneration * **Validate with A/B generation** (2 outputs) before committing to a larger batch * **Use presets** to apply consistent styling without extensive prompt iteration — fewer failed generations from style inconsistency * **Select from history** instead of re-uploading reference images — keeps workflows clean and avoids accidental duplicate uploads *** #### Credit Usage by Configuration Relative cost comparison across common configurations: | Configuration | Relative Cost | | ------------------------------------ | ------------- | | ImageGen Fast + 720p + 1 generation | Lowest | | Flux Dev + 1K + 1 generation | Very Low | | Nano Banana 2 + 1K + 1 generation | Low–Moderate | | Nano Banana 2 + 2K + 2 generations | Moderate | | Nano Banana Pro + 2K + 2 generations | High | | Nano Banana Pro + 4K + 4 generations | Highest | > **Note:** Exact credit costs vary by model as listed above. This table reflects relative cost relationships — not absolute values. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://docs.qolaba.ai/image-generation/understanding-credits-usage.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.