# What Flow AI can make > Content revision: 2026-11-09 Create from a prompt or your own media, then choose a model and settings that fit the result you want. Flow AI brings video, image, music and audio tools into one place. ## Video modes ### Image to video Upload a photo and describe the motion, from a small portrait movement to a product reveal. The selected model determines the available length and output size. 16 credit-based models support this mode: Veo 3.1 Lite, Veo 3.1 Basic, Veo 3.1 Premium, Seedance 2.5, Seedance 2.0, Seedance 2.0 Fast, Seedance 2.0 Mini, MiniMax H3, Google Omni Lite, Google Omni, HappyHorse 1.0, Kling 3.0, Wan 2.6, Grok, Kling 2.6, Wan 2.5. Veo 3.1 Free is the free option for this mode; see [free creation](https://flowai.im/pricing.md#starting-for-free). - [Image to video](https://flowai.im/image-to-video) ### Text to video Describe a subject, setting, action and camera movement to create a scene without a source photo. 16 credit-based models support this mode: Veo 3.1 Lite, Veo 3.1 Basic, Veo 3.1 Premium, Seedance 2.5, Seedance 2.0, Seedance 2.0 Fast, Seedance 2.0 Mini, MiniMax H3, Google Omni Lite, Google Omni, HappyHorse 1.0, Kling 3.0, Wan 2.6, Grok, Kling 2.6, Wan 2.5. Veo 3.1 Free is the free option for this mode; see [free creation](https://flowai.im/pricing.md#starting-for-free). - [Text to video](https://flowai.im/text-to-video) ### Start and end frame Provide a first image and a last image, then describe the transition between them. Select the frame-pair mode before adding both pictures. 10 credit-based models support this mode: Veo 3.1 Lite, Veo 3.1 Basic, Veo 3.1 Premium, Seedance 2.5, Seedance 2.0, Seedance 2.0 Fast, Seedance 2.0 Mini, MiniMax H3, Google Omni Lite, Kling 3.0. Veo 3.1 Free is the free option for this mode; see [free creation](https://flowai.im/pricing.md#starting-for-free). - [Start and end frame](https://flowai.im/create/video) ### Reference to video Supply images, and where supported video or audio, to guide a new scene. Use the [video workspace](https://flowai.im/create/video) for Veo and Google Omni Lite image references, or the [multi-reference tool](https://flowai.im/multi-reference) for Seedance 2.0, Fast, Mini, Seedance 2.5, MiniMax H3, Google Omni or HappyHorse. 11 credit-based models support this mode: Veo 3.1 Lite, Veo 3.1 Basic, Veo 3.1 Premium, Seedance 2.5, Seedance 2.0, Seedance 2.0 Fast, Seedance 2.0 Mini, MiniMax H3, Google Omni Lite, Google Omni, HappyHorse 1.0. Veo 3.1 Free is the free option for this mode; see [free creation](https://flowai.im/pricing.md#starting-for-free). - [Reference to video](https://flowai.im/create/video) ## Image modes ### Text to image Describe a product, character, setting or illustration and choose the composition you need. 12 credit-based models support this mode: Nano Banana 2, Nano Banana Pro, Nano Banana, GPT Image 2, Seedream 5, Seedream 4.5, Flux 2 Flex, Flux 2 Pro, Wan 2.7, Wan 2.7 Pro, Grok Imagine, Z-Image. Nano Banana 2 Free is the free option for this mode; see [free creation](https://flowai.im/pricing.md#starting-for-free). - [Text to image](https://flowai.im/text-to-image) ### Image to image Upload references and describe what should change, such as the background, styling or composition. Reference count and output size depend on the model. 11 credit-based models support this mode: Nano Banana 2, Nano Banana Pro, Nano Banana, GPT Image 2, Seedream 5, Seedream 4.5, Flux 2 Flex, Flux 2 Pro, Wan 2.7, Wan 2.7 Pro, Grok Imagine. Nano Banana 2 Free is the free option for this mode; see [free creation](https://flowai.im/pricing.md#starting-for-free). - [Image to image](https://flowai.im/image-to-image) ## Music modes ### Text to music Describe the style and mood, add lyrics for a song or select instrumental output. 1 model supports this mode: Music V6. - [Text to music](https://flowai.im/create/music) ### Audio to music Upload one audio reference and describe the cover or musical direction. 1 model supports this mode: Music V6; references must be 6–300 seconds and at most 50 MiB. - [Audio to music](https://flowai.im/create/music) ## Audio modes ### Text to audio Write a voiceover, dialogue or sound-scene brief. 1 model supports this mode: Seed Audio 1.0, with optional pitch, speed and volume controls. - [Text to audio](https://flowai.im/create/audio) ### Reference-guided audio Add up to three audio references to a text brief to guide the sound. Seed Audio 1.0 supports this workflow; each reference can be at most 30 seconds and 10 MiB. - [Reference-guided audio](https://flowai.im/create/audio) ## Editing tools ### Video editing and extension Video Edit uses Seedance 2.5 or MiniMax H3 with a required source clip, an edit prompt and up to nine optional image references. Seedance 2.5 accepts 2–30-second source clips and 4–30-second output; MiniMax H3 accepts 2–15-second sources and 4–15-second output. Video Extend accepts a 1.625–60-second source clip and up to three optional audio references, each 2–15 seconds with 15 seconds total. Choose an added duration of 0.71–15 seconds and either the continuation alone or the extended clip. Both tools quote the selected task before submission. - [Video edit](https://flowai.im/video-edit) - [Video extend](https://flowai.im/video-extend) ### Motion control and lip sync Motion Control uses a JPEG/PNG character image and a 3–30-second MP4/MOV motion reference, with 720p or 1080p quality. Lip Sync uses a JPEG/PNG/WebP portrait and 5–30-second audio, with 480p, 768p, 1080p or 2K output settings. Each tool shows its upload requirements and includes sample materials you can choose or replace. - [Motion control](https://flowai.im/motion-control) - [Lip sync](https://flowai.im/lip-sync) ### Image editing and enhancement Edit an existing picture, enlarge it, remove a background or extend its canvas. Use the watermark-removal tool only with images you have the right to edit. - [Image editor](https://flowai.im/image-editor) - [Image upscaler](https://flowai.im/ai-tools/image-upscale) - [Background remover](https://flowai.im/ai-tools/remove-background) - [Image expand](https://flowai.im/ai-tools/image-expand) - [Watermark remover](https://flowai.im/ai-tools/watermark-remover) ### Video enhancement Upload a clip to the dedicated video upscaler and select one of its supported enhancement settings. - [Video upscaler](https://flowai.im/ai-tools/video-upscaler) ## Choosing a workflow | Starting material | Goal | Recommended entry | Next step | | --- | --- | --- | --- | | A written scene | Make a short video | [Text to video](https://flowai.im/text-to-video) | Choose a model, set its supported output options and review the quote. | | One photo | Animate a person, product or place | [Image to video](https://flowai.im/image-to-video) | Upload the image and describe one action or camera movement. | | A first and a last image | Control a transition | [Video workspace](https://flowai.im/create/video) | Choose a compatible model and Start and end frame mode, then assign both images. | | Several images or supported video/audio references | Guide a consistent new scene | [Video workspace](https://flowai.im/create/video) | Select reference mode and a compatible model. Use the multi-reference tool for Seedance, MiniMax H3, Google Omni or HappyHorse. | | An existing video | Change or continue the shot | [Video edit](https://flowai.im/video-edit) or [Video extend](https://flowai.im/video-extend) | Upload the clip, describe the edit or continuation and review the tool's settings. | | A character photo and motion clip | Reuse a movement | [Motion control](https://flowai.im/motion-control) | Set the character image and motion reference, then submit when ready. | | A visual idea or reference pictures | Create or restyle images | [Image workspace](https://flowai.im/create/image) | Choose text-to-image or image-to-image and a suitable model. | | A music brief, lyrics or audio reference | Make a song, instrumental or cover | [Music](https://flowai.im/create/music) | Set the style, vocal/instrumental choice and supported reference inputs. | | A script or sound-scene brief | Create speech, dialogue or ambience | [Audio](https://flowai.im/create/audio) or [Speech workspace](https://flowai.im/text-to-speech) | Use Seed Audio 1.0, optional audio references and voice controls; choose the speech workspace for a focused voiceover or dialogue project. | | A story brief and optional visual references | Plan several shots | [Storyboard](https://flowai.im/storyboard) | Review the storyboard image and video brief before continuing to a separately quoted video task. | | A product or multi-step visual brief | Build connected outputs | [Canvas](https://flowai.im/canvas) | Start a project or template, connect compatible inputs and run the needed generation steps. | ## Inputs, settings and results 1. Choose a tool and model for the result you want, from an animated photo to a song or product image. 2. Add a prompt and the required media. Assign the first and last images for a frame transition, or upload references that guide the subject, style and movement. 3. Choose the available aspect ratio, quality, duration and sound options. Check the upload requirements and any plan requirements for your selection. 4. Review the cost before generating. Free models need no credits or sign-up and take longer to generate. For credit-based creation, sign in and check the selected model's [credit and plan requirements](https://flowai.im/pricing.md#access-conditions). 5. Preview the result and download a copy. Sign in to keep creations with your account; download guest results or save them before they expire. 6. Build on the result with the available actions, such as editing an image, making a video from it or enhancing a clip. Review the next step's inputs and cost before starting. ## Storyboard workflows [Storyboard](https://flowai.im/storyboard) turns a story or campaign brief into a storyboard image and an editable video prompt. It prepares a scene sequence; generating a final video is a separate step. 1. Describe the story or product sequence and optionally add up to four reference images. 2. Choose 9 or 21 panels, a target ratio of 16:9, 9:16 or Auto, and a planned video length of 5, 10 or 15 seconds. The length describes the video brief, not an already generated clip. 3. Sign in, review the storyboard generation cost and submit. The workflow plans the panels, generates the storyboard image and prepares a video prompt. 4. Inspect or download the storyboard image. Edit, copy or save the video prompt to refine pacing, shots and continuity. 5. Choose Continue to video to continue in the [reference-video workspace](https://flowai.im/multi-reference). Review the storyboard, model, settings and new credit cost before submitting the video generation. ## Canvas workflows [Canvas](https://flowai.im/canvas) brings ideas, references and generated media together in a visual workspace. Build a connected creative process, save it as a project and return to it when you want to develop another version. - Start a new project or preview a [product launch template](https://flowai.im/canvas/examples/prism-speaker-launch). Make your own copy to adapt it to your product or campaign. - Arrange prompts, notes, pictures, videos and audio. Use notes to organize ideas and connect the materials you want to guide the next creative step. - Choose a model and settings for each generation, review the credit cost and create the result. Use compatible images, clips or audio as references for the next step. - Preview and download the outputs you need. Name and save the project so you can reopen it from Canvas. - Import or export a project to keep a copy of your workflow. Download important media separately. Sign in to create or copy projects, save your work and generate. The model and settings you choose determine the generation cost and plan requirements. ## Templates and samples Choose motion references, portrait and audio samples, or prepared Canvas projects as a starting point. Replace the sample materials with your own and adapt the creative direction to your project. - [Templates and samples](https://flowai.im/tools.md#templates) ## Failures and output use Failed-task credit restoration follows the relevant workflow; contact support if a failed task has not been reconciled. A restored generation balance is distinct from a payment refund. See [pricing and refunds](https://flowai.im/pricing.md) for the complete distinction. Personal or commercial use depends on the plan, applicable law and third-party rights, as explained in the [Terms](https://flowai.im/terms-of-service). Review outputs before publishing; generated media is not automatically cleared for every use. ## Official sources - [Video generator](https://flowai.im/video-generator) - [Image generator](https://flowai.im/image-generator) - [Music generator](https://flowai.im/music-generator) - [Audio workspace](https://flowai.im/create/audio) - [Model reference](https://flowai.im/models.md) - [Terms of Service](https://flowai.im/terms-of-service)