ChatCut’s video model selector gives each request an explicit generation model. The available inputs and settings change when you switch models, so choose the model before attaching reference media.
Video generation requires Pro and also consumes credits. See Plans and Credits for access and Credits Policy for the current model rates.
Where is the model selector?
There are two ways to select a model in the editor:
- In Video Gen mode, select the model name in the toolbar below the prompt. The menu lists Seedance 2.5, Seedance 2.0, Seedance 2.0 Fast, Seedance 2.0 mini, Kling 3.0, and Gemini Omni.
- In Agent mode, select the box icon labeled Video generation model below the message field, then choose a model. It is attached to the current AI draft; describe the video task and send the request.
For direct control over inputs, duration, ratio, resolution, and output format, use Video Gen mode.
Choose a model before configuring the request
Use Compare Video Model Capabilities for task-based guidance. Use the Model Capability Matrix for current input counts, editing, extension, and output tiers. Keeping those changing limits in the reference table prevents this operational page from carrying a second copy.
How do I choose the input type and output settings?
- Switch the AI composer to Video Gen.
- Select the model name in the bottom toolbar.
- Select the sliders icon labeled Video settings.
- Under Input, choose the operation supported by that model.
- Set Aspect Ratio, Resolution, and Duration where available. Seedance 2.5 also offers Output format with MP4 or MOV.
- Add media through the visible First Frame, Last Frame, Reference, Image Ref, or source-video slot.
- Write the prompt and select Send message.
When a frame or source clip determines the result’s dimensions or duration, ChatCut locks or marks the inherited setting instead of pretending it can be changed.
How do I edit or extend a video?
Seedance 2.5
Open Video settings and choose Edit video or Extend video under Input. Add a video from the project library, describe the edit or extension, and send the request. Edit video accepts one 4–30 second source and matches its duration. The editing and extension flows default to MOV output, which you can review in the settings panel.
Gemini Omni
Select Gemini Omni, open Video settings, and choose Edit video. Drag one uploaded project video into the source slot labeled Drop a video from your library, then describe only the changes you want. Sources longer than roughly 10 seconds are rejected. Ratio, length, and resolution are inherited from the source while editing.
What happens to incompatible inputs when I switch models?
ChatCut keeps only media the new model can accept. For example, switching to Kling removes audio and video references; switching to Gemini Omni changes the composer to one of its mutually exclusive input modes. Recheck the visible input slots after changing models before you send.
The credit estimate next to the send button updates with the selected model, duration, resolution, and eligible video inputs. Treat the estimate and any preflight dialog in the editor as the current billing source of truth. For the current per-second model table and Seedance 2.5 input billing, see Credits.