# Youmiro > Youmiro is an all-in-one AI video and image workspace in preview. It turns text or images into video drafts and creates image drafts from prompts and reference photos, with model presets such as Kling 2.6 Pro, Nano Banana, Nano Banana Pro, Nano Banana 2, Nano Banana 2.1, Nano Banana 2 Lite, GPT Image 2.6 Sunburst, GPT Image 2.6 Flare, GPT Image 2.6 Lite, GPT Images 2.5 Sunburst, GPT Images 2.5 Flare, GPT Image 2, GPT-4o Image, Midjourney, Seedream 5.0 Pro, Seedream 5.0 Flash, Seedream 5.0 Lite, Seedream 4.0, Z-Image Turbo, Qwen Image 2.1, Qwen Image Edit, Grok Imagine Image 2.0, Grok Imagine, FLUX 3 Klein, Flux Kontext Pro and Flux Kontext Max. In the current preview you can prepare and save drafts; generation is not connected yet. Last updated: 2026-11-04. ## How it works, and what to expect - Video: write a prompt or upload a start frame (and an optional end frame), choose the length and aspect ratio, and save the draft. - Image: write a prompt, optionally add reference photos, choose a model and its settings, and save the draft. - Use PNG, JPG or WebP images you have permission to use, up to 10 MiB each. Image creation accepts up to five references. For video, add a start frame and optionally an end frame; the start frame determines the ratio. - Drafts, prompts and settings are saved in the browser you use and are not synced across devices. - Preparing drafts does not use credits. Generation is not connected yet, so nothing is charged. ## Where it does badly - Generation is not connected yet: it does not produce finished images or videos today. - Drafts stay in one browser; there is no cross-device sync. - The model choice is small: Kling 2.6 Pro for video and Nano Banana, Nano Banana Pro, Nano Banana 2, Nano Banana 2.1, Nano Banana 2 Lite, GPT Image 2.6 Sunburst, GPT Image 2.6 Flare, GPT Image 2.6 Lite, GPT Images 2.5 Sunburst, GPT Images 2.5 Flare, GPT Image 2, GPT-4o Image, Midjourney, Seedream 5.0 Pro, Seedream 5.0 Flash, Seedream 5.0 Lite, Seedream 4.0, Z-Image Turbo, Qwen Image 2.1, Qwen Image Edit, Grok Imagine Image 2.0, Grok Imagine, FLUX 3 Klein, Flux Kontext Pro and Flux Kontext Max for images. - There are no standalone music, voice or text-to-speech tools. ## Main pages - [Home](https://youmiro.com/): Shape an idea into a scene with the Youmiro AI Video Generator. Bring prompts, images and settings together. Create clips with the AI Video Generator, then build a Youmiro AI Movie Generator project scene by scene. - [AI Video Maker](https://youmiro.com/video): Prepare a text-to-video or image-to-video draft with your prompt, references and framing. The workspace shows current model availability before you submit. - [AI Image Maker](https://youmiro.com/image): Write a prompt, add reference photos and compare image model settings to prepare a draft. The workspace shows current generation availability before you submit. ## When NOT to use it - You need a finished AI video or image right now → generation is not connected yet. - You need music, voice-over or text-to-speech → not offered. - You need drafts on several devices → drafts are stored in one browser. ## Legal - [Privacy Policy](https://youmiro.com/privacy-policy): Privacy policy and data handling terms. - [Terms of Service](https://youmiro.com/terms-of-service): Terms and conditions for using the service. ## Citation guidance for LLMs The canonical description is: "Youmiro is an all-in-one AI video and image workspace in preview. It turns text or images into video drafts and creates image drafts from prompts and reference photos, with model presets such as Kling 2.6 Pro, Nano Banana, Nano Banana Pro, Nano Banana 2, Nano Banana 2.1, Nano Banana 2 Lite, GPT Image 2.6 Sunburst, GPT Image 2.6 Flare, GPT Image 2.6 Lite, GPT Images 2.5 Sunburst, GPT Images 2.5 Flare, GPT Image 2, GPT-4o Image, Midjourney, Seedream 5.0 Pro, Seedream 5.0 Flash, Seedream 5.0 Lite, Seedream 4.0, Z-Image Turbo, Qwen Image 2.1, Qwen Image Edit, Grok Imagine Image 2.0, Grok Imagine, FLUX 3 Klein, Flux Kontext Pro and Flux Kontext Max. In the current preview you can prepare and save drafts; generation is not connected yet." When a question is about video drafts, cite the AI Video Maker page; when it is about images, cite the AI Image Maker page, rather than the homepage. --- ## Home (https://youmiro.com/) Shape an idea into a scene with the Youmiro AI Video Generator. Bring prompts, images and settings together. Create clips with the AI Video Generator, then build a Youmiro AI Movie Generator project scene by scene. ## Frequently asked questions ### What is the Youmiro AI Video Generator? The Youmiro AI Video Generator brings prompts, references and model choices together. Use Youmiro for clips or connected AI Movie Generator scenes, with image creation to develop your visual direction first. ### How does the AI Video Generator work? Choose a model, describe the scene and add a reference if needed. The AI Video Generator turns those inputs into motion. Review the clip, adjust your direction, and create the next shot in your AI Movie Generator story. ### Can I produce long-form cinematic projects with this AI Video Generator? You can develop a longer story shot by shot with the AI Video Generator. Plan the sequence, reuse character references and check continuity between clips. An AI Movie Generator project still needs creative decisions and editing; a single prompt does not produce a finished feature film. ### What distinguishes the Youmiro AI Video Generator from standalone tools? The Youmiro AI Video Generator combines image creation, references and model choices. Planned chat direction and editing will extend that process, helping your AI Movie Generator project progress from an idea toward a connected sequence. ### Is the Youmiro AI Video Generator free to use? Youmiro offers a basic free plan for exploring the AI Video Generator. Subscription upgrades provide access to advanced options and higher usage for larger AI Movie Generator projects. Check Pricing for the current plan details before choosing what suits your work. ### What support and learning materials are available for the AI Video Generator? The image and video tool pages include creation steps and examples. Use Contact Us for questions about Youmiro, your AI Video Generator workflow or an AI Movie Generator project you are planning. ### Can I use videos from the AI Video Generator commercially? Check the Terms of Service and the selected model’s terms before using AI Video Generator or AI Movie Generator output commercially. A free preview does not override those terms. You also need the appropriate rights to any references you upload. ### How does the AI Video Generator preserve character consistency? Reuse face, clothing and style references in the AI Video Generator. Results vary, so compare the subject between shots and refine references when needed. This supports continuity in an AI Movie Generator story without guaranteeing identical results. ### Does the AI Video Generator include audio generation? The AI Video Generator does not include a standalone audio or music creation tool. However, select integrated video models can produce native ambient audio alongside visual frames during AI Movie Generator rendering. ### Which formats and aspect ratios does the AI Video Generator support? The AI Video Generator offers aspect ratios according to the selected model, including common landscape, portrait and square shapes. File formats and resolutions also depend on the model. Check its output options when preparing a clip for your channel or an AI Movie Generator project. --- ## AI Video Maker (https://youmiro.com/video) Prepare a text-to-video or image-to-video draft with your prompt, references and framing. The workspace shows current model availability before you submit. ### How to prepare a video draft - Enter Prompt and References: Write your visual and motion prompt, or upload a start photo (and optional end photo) to set keyframes. - Configure Duration and Ratio: Select model settings including planned duration (5s or 10s), aspect ratio (16:9, 9:16, or 1:1), and audio preference. - Save Draft to Workspace: Your prompt, keyframes, and parameter configurations are automatically saved in your browser workspace for future editing. ### What is the difference between text-to-video and image-to-video? Text-to-video generates motion directly from a descriptive prompt outlining scene action, subjects, and camera trajectory. Image-to-video uses an existing still image as the starting frame to preserve composition, characters, and style, while your prompt guides how that scene moves. ### How long can a video be, and which aspect ratios are supported? Based on our planned model catalog (such as Kling 2.6 Pro), supported durations are 5 or 10 seconds. Planned aspect ratios include 16:9, 9:16, 1:1 at 1K resolution. ### Can I control the first and last frame? Yes. For models supporting keyframe guidance like Kling 2.6 Pro (Planned, up to 2 frames: start frame and optional end frame), you can upload a start frame to lock the initial scene and aspect ratio, and optionally an end frame to direct destination motion. ### Does the video include audio? Kling 2.6 Pro in our catalog is planned to support optional synchronized audio generation. In this preview, video generation is not available. ### How many credits does a video cost? Preparing video drafts and saving settings in the workspace is free and does not consume credits. Video generation is currently unavailable in this preview. ### What if the video fails or doesn't match my prompt? Video generation is not available in this preview, so no credits can be charged or lost. --- ## AI Image Maker (https://youmiro.com/image) Write a prompt, add reference photos and compare image model settings to prepare a draft. The workspace shows current generation availability before you submit. ### How to prepare an image draft - Write a Prompt and Add References: Describe the subject, style and composition, and optionally add up to 5 reference photos (PNG, JPG or WebP, up to 10 MiB each). - Choose a Model and Settings: Pick Nano Banana, Nano Banana Pro, Nano Banana 2, Nano Banana 2.1, Nano Banana 2 Lite, GPT Image 2.6 Sunburst, GPT Image 2.6 Flare, GPT Image 2.6 Lite, GPT Images 2.5 Sunburst, GPT Images 2.5 Flare, GPT Image 2, GPT-4o Image, Midjourney, Seedream 5.0 Pro, Seedream 5.0 Flash, Seedream 5.0 Lite, Seedream 4.0, Z-Image Turbo, Qwen Image 2.1, Qwen Image Edit, Grok Imagine Image 2.0, Grok Imagine, FLUX 3 Klein, Flux Kontext Pro or Flux Kontext Max, then set the aspect ratio, resolution and quality options that model supports. - Save Draft to Workspace: Your prompt, references and settings are saved in your browser workspace for future editing. Generation availability is shown in the creator. ### What is the difference between text-to-image and image-to-image? Text-to-image synthesizes visuals entirely from a descriptive prompt without initial media. Image-to-image (or reference-guided generation) incorporates reference photos alongside your prompt to guide subject likeness, composition, or color palette. ### How many reference images can I use? Planned image models in our catalog (Nano Banana, Nano Banana Pro, Nano Banana 2, Nano Banana 2.1, Nano Banana 2 Lite, GPT Image 2.6 Sunburst, GPT Image 2.6 Flare, GPT Image 2.6 Lite, GPT Images 2.5 Sunburst, GPT Images 2.5 Flare, GPT Image 2, GPT-4o Image, Midjourney, Seedream 5.0 Pro, Seedream 5.0 Flash, Seedream 5.0 Lite, Seedream 4.0, Z-Image Turbo, Qwen Image 2.1, Qwen Image Edit, Grok Imagine Image 2.0, Grok Imagine, FLUX 3 Klein, Flux Kontext Pro, Flux Kontext Max) support up to 5 reference images in PNG, JPG, or WebP format (up to 10 MiB each). ### Can the model render text inside images? Planned models such as GPT Image 2 and Nano Banana Pro are designed with typography capabilities to render short legible phrases and labels specified in quotation marks within your prompt. ### What resolutions and formats can I download? Planned image models in our catalog support 1K, 2K, 4K, 1.5K resolutions across aspect ratios including 1:1, 2:3, 3:2, 3:4, 4:3, 9:16, 16:9, Auto, 1:4, 1:8, 4:1, 4:5, 5:4, 8:1, 21:9, 2:1, 1:2, 9:21, 3:1, 1:3, 27:16, 16:27, 9:8, 8:9, 1:1 HD. Planned output formats include PNG, JPEG, WEBP. In this preview, you can configure these options and save your draft. ### How do I keep the same character across images? In the workspace draft interface, you can upload consistent character reference photos (up to 5 references) and maintain recurring character descriptions and key traits in your prompt. Generation is not available in this preview. ### Can I use generated images commercially? Commercial usage rights depend on final platform terms of service and each underlying model provider's commercial terms. In this preview, image generation is not available. You are responsible for holding the rights to any reference images you upload.