Skip to main content

Overview

The Creative Studio (/studio) is the main creative workspace in DAWSEN AI. It uses a chat-based interface where each message generates images or videos, building a visual conversation you can iterate on. Multiple sessions let you organize work by product or campaign.

How It Works

1

Open Studio

Navigate to Studio from the main navigation.
2

Start or Select a Session

Create a new session for each product or project. Sessions keep your generation history organized and separate.
3

Describe What You Want

Type a prompt in the input bar. Choose between Image or Video output mode.
4

Generate

Click Generate. Results appear in the chat thread below your prompt.
5

Iterate

Use any generated image as a reference for the next generation, edit with inpainting, remove its background, or turn it into a video — all without leaving the Studio.

Sessions

Sessions are named workspaces that group your generations together.
  • Create up to the session limit for your plan
  • Rename sessions to match your product or campaign
  • Switch between sessions from the left sidebar
  • Delete sessions you no longer need

Image Generation

Models

Choose from multiple AI models in the toolbar:
ModelDescription
Nano BananaFast, high-quality generations
GPT-4oOpenAI’s image model
FLUX KontextExcellent for reference-guided generation
SeedreamStylized, creative outputs

Settings

  • Aspect Ratio: 10 options from 1:1 to 21:9
  • Resolution: 1K, 2K, or 4K
  • Variants: Generate 1–4 images at once
  • Visual Style: Choose from 100+ curated styles

Reference Images

Attach up to 5 reference images to guide the generation:
  • Upload from your device
  • Select from your Gallery
  • Use any previously generated image as a reference

Prompt History

Click the clock icon to see your 10 most recent successful prompts and reload them with one click.

Video Generation

Switch to Video mode in the input bar to generate videos.

Text-to-Video

Describe the video you want and select a model:
ModelNotes
Kling 2.6High quality, recommended
Kling 2.5Fast generation
Veo 3Google’s video model
Wan 2.6Strong motion control
Sora 2 ProOpenAI’s video model
Settings: duration (5s or 10s), aspect ratio, sound on/off.

Image-to-Video

Set a Start Frame image to animate from a specific product photo. Optionally set an End Frame for first-last-frame interpolation — the AI fills in the motion between the two frames.

Reference Video

Attach a reference video for motion control — the AI matches the motion style of your reference.

Editing Tools

Inpainting

Click the edit (mask) icon on any generated image to open the inpainting editor. Paint over the area you want to change, write a prompt, and regenerate just that region.

Background Removal

Click the background removal icon on any image to instantly remove its background. The result is saved to your Gallery as a transparent PNG.

Chat Actions

Hover over any generated image in the chat to access:
  • Use as Reference — adds the image to the next generation’s reference images
  • Edit with Mask — opens inpainting editor
  • Remove Background — removes background instantly
  • Download — saves the full-resolution file
  • Create Video — opens the video generation modal with this image as the start frame
  • Delete — removes the image from the session

Credit Costs

ActionCost
Image (1K)1 credit
Image (2K)2 credits
Image (4K)3 credits
Background Removal1 credit per image
Video (5s)8 credits
Video (10s)16 credits

Video Generation

Learn more about standalone video generation with all 5 models.