Animanager AI

Generate next to your shots.Pay only for what you use.

Each generation shows its price before it runs. Follow one shot of the fictional short "Greenhouse", from references to a reviewed take.

AI inside the pipeline.

Image and video models on the canvas

Nano Banana, GPT Image, Seedream, Seedance, MiniMax, Kling and Veo. Switch models on the same node.

Workflows, not one-off prompts

Wire references and earlier results into a generator. Change one and run it again.

Results go into production

Publish a take as a revision and review it frame by frame.

Enhance, clean and upscale

Upscale, remove noise, add sound effects or turn an image into a 3D model.

Assistants that know the canvas

Claude, Codex or another MCP client can place media and prepare generators for you.

See the MCP

From reference to reviewed take.

Illustrations of the real interface. The robot, the greenhouse and the takes were made with these tools: GPT Image 2.5, MiniMax H3 Max and Topaz.

Step 1 · Assets

Define the cast and the set once.

Save the character views, the location and the props as assets. Every shot uses the same references.

  1. Build the identity

    Front, side, three-quarter and back views of the robot in one asset.

  2. Find it from any shot

    Search assets from the shot, the sequence or the whole project.

  3. Drag it onto the canvas

    Drop it on the canvas, ready to wire into a generator.

Illustrative Animanager Assets panel beside a shot canvas in the blue theme, based on the real interface with fictional Greenhouse project data.

Step 2 · Generate

Wire the references, explore the light.

Connect the robot and the greenhouse to an image generator, write a short prompt and compare the results.

  1. References in, results out

    Green wires show what fed each result. Change a reference and run it again.

  2. Choose the model per node

    Switch models without rebuilding the node.

  3. Keep the direction

    Mark the chosen frame so the team knows which one to continue.

Illustrative Animanager canvas with references wired into an image generator and three results in the blue theme, based on the real interface with fictional Greenhouse project data.

One prompt, one robot, three lighting directions.

Sunrise
Rain
Night

Step 3 · Animate

Turn the chosen frame into a take.

Use the chosen keyframe as the first frame of a video and describe the action with timings. The robot, the set and the light stay the same.

  1. Start from the frame

    The first frame sets the look, so the take matches the keyframe.

  2. Describe the action in seconds

    From 0 to 2 seconds it rolls forward, from 2 to 4 it tilts its head. Short clips, quick tries.

  3. Try the other lights

    Run the same prompt on the rain and night frames.

0:00/0:00
0:00/0:00
0:00/0:00

Step 4 · Publish

Publish the take into the production.

The chosen take becomes a revision of the shot, with its stage, status and chat.

  1. A revision like any other

    It shows the files, the preview and who published it, tagged as generated.

  2. Same conversation

    The lead answers on the stage. The next take starts from that note.

  3. Same files

    Get files brings the keyframe and the take to your computer for the edit.

Illustrative Animanager stage conversation with a generated revision in the blue theme, based on the real interface with fictional Greenhouse project data.

Step 5 · Review

Review it like any other shot.

Play the take frame by frame and draw on the frame that needs a change. Clients can review through a link, without a seat.

  1. Frame-accurate notes

    "Slow the head tilt." The note sits on frame 72, not in an email.

  2. Compare takes

    Put the sunrise and rain takes side by side.

  3. Clean up and deliver

    Upscale or clean the chosen take, then deliver it from the project.

Illustrative Animanager frame-by-frame review of a generated take in the blue theme, based on the real interface with fictional Greenhouse project data.

See it in a real pipeline