English
AI motion design in Blender
Create short 3D clips, product orbits and text animation with ChatGPT, Claude, Codex or Qwen Code — describe the motion in words and refine the scene in chat. Blender Bridge connects your AI to your Blender: it animates objects, sets up the camera and renders the video.
From a still to a moving shot
For a product clip, start with the subject, composition, materials and lighting. Once you choose the look, add movement: turn the object, orbit the camera, move in toward a detail or introduce text. The AI can work with Blender's keyframes and animation curves. Ask for constant speed, gradual acceleration and deceleration, or distinct stages of an action.
Set the duration, frame rate and format, such as a vertical social video or a horizontal shot for your website. If you need repeating animation, explicitly ask for a loop.
Four prompts for a clip
- "Create a simple bottle on a dark background and make an eight-second camera orbit at 30 frames per second."
- "Animate this box opening: raise the lid, then slowly move the camera toward the object inside."
- "Make a 1080 × 1920 vertical clip: rotate the product against a dark background, then reveal its name at the end."
- "Run a background render of the selected frame range to MP4. Report progress, save the finished video and show me where to get the file."
For the last prompt, specify the frame range or ask to use the scene's range.
The workflow
You install the Blender Bridge app and connect your AI once. From then on, you create and change the scene in plain words in the chat — no code required.
- Connect your AI, open a scene or describe the subject and shot.
- Ask it to set up lighting and a camera, then add movement with the duration you want.
- Review the animation in Blender. Ask for screenshots of key moments or a short preview render.
- Refine the speed, path, focus and framing. For a loop, check the transition from the end back to the start.
- Ask it to save the
.blendfile and run a background render of the final range in your chosen format.
Blender Bridge saves checkpoints of the AI's work. Return to an available checkpoint to try a different version of the movement.
Background video rendering
Explicitly request Blender Bridge background rendering. It starts a copy of the scene in a separate Blender process. The main Blender stays responsive, progress is available, and you can stop the render from chat. That render uses the launch-time scene copy. Later edits to the open scene belong to your next version.
Before rendering the full clip, test a short range with the same lighting, materials and resolution. Request an MP4 video or an image sequence for editing elsewhere. Ask for the completed video file after rendering finishes.
Where the files go
The video or image frames are saved in your project folder. With a server connection, download the completed result from chat for use in your film, on your website or on social media. The saved .blend scene keeps the objects, cameras and animation available for your next revision.
Blender Bridge MCP connects to Codex, Claude Code, Qwen Code, Kimi Code and other MCP clients. Use a server connection for ChatGPT and Claude. The browser extension works with ChatGPT, Claude, Gemini and DeepSeek on the computer running Blender.
Explore animation, rendering, product visualization and model exports.
Next: Learn Blender with AI: build scenes and learn through your own project