Live Nov 18 · Cohort 3Nov 18, 19:00 MSK — free hands-on AI agents live workshop · Live Cohort 3 starts Nov 23Join Nov 18 Live →Cohort 3 (Nov 23)

Dreamina Launches Motion Studio: Motion Graphics Written as Code From a Prompt

Dreamina AI has released Motion Studio. From a text description, the service writes animation code and renders a clip from it, rather than drawing frames with a video model. It assembles ideas, text and images into a script, and you can edit layers, element entry points and timings precisely. Similar services already exist from Claude, Hera, Jitter and GenMotion. For teams, it is a way to make intros and explainer videos without After Effects.

AI-processed from TG: @cgevent (Метаверсище и ИИще); edited by Hamidun News
Dreamina Launches Motion Studio: Motion Graphics Written as Code From a Prompt
Source: TG: @cgevent (Метаверсище и ИИще). Collage: Hamidun News.
Listen to article
Article Fact Sheet & Key Figures
ParameterDetails
Primary sourceTG: @cgevent (Метаверсище и ИИще) ↗
Editor-in-Chief
Claude Motionbeta only for Team and Enterprise, outputs MP4
Key metric #2Hera and Jitter are layer-based editors; Hera has a free tier
Engines under the hoodRemotion, Motion Canvas, HyperFrames, givemotion, Manim

Dreamina AI has released Motion Studio, a tool that makes motion graphics by writing code rather than generating pixels. You describe a clip in words, and the system arranges your ideas, text and media files into a script with carefully tuned pacing, then renders the result. It sounds like yet another video generator, but it works differently, and for practitioners that difference matters.

Why code instead of a video model

A typical video model redraws every frame from scratch. If an intro has a typo in the headline or a wrong number on an infographic, you have to redo the whole clip, and the result comes out slightly different each time. In the code-rendered approach, the animation is a program: it has layers, points where elements appear, and timings. Want to shift a headline by half a second or change a number? You edit a parameter instead of regenerating everything. Motion Studio adds ready-made style presets and refinement through follow-up prompts.

The author of the Telegram channel «Метаверсище и ИИще» notes that Dreamina itself has little to boast about here. It is interesting mainly because the creators of Seedance are behind it. The trend itself, in his words, is growing: Claude and other models write good code for Three.js and similar engines, so a whole wave of such services should be expected.

Who else is playing in this space

Claude Motion is the closest analogue. You describe a clip in words, Claude writes the animation as code, and you can then change the text, numbers and motion and export MP4. It does not use a video generation model. The limitation: it is still a beta for Team and Enterprise plans and is not available on the regular Pro plan.

Hera is an AI motion graphics editor with editable layers. It makes ad intros, infographics, animated headlines and product videos. You refine the result with prompts or edit individual elements by hand. It is more a graphics editor with AI than an environment for programmers. It has a free tier.

Jitter is a browser-based editor that resembles a simplified After Effects. AI Superagents change layers, composition and transitions from a prompt. It suits people who want to control animation on a timeline and not look at code.

GenMotion is a local environment where Claude Code or Codex builds animated scenes and you adjust them in plain language. It handles clips, titles, voiceover and sound effects, and exports to MP4, WebM and GIF. The limitation: the current version on the site is for macOS with Apple Silicon only, with no Windows version.

Engines you can call from agents

If you don't need a ready-made service, there are engines underneath, and you can call them directly or from your own agents:

  • Remotion turns React/TypeScript into video. A versatile option for motion graphics, ads and automation.
  • Motion Canvas uses TypeScript for vector animation. Good for explainer videos, diagrams and syncing with voiceover.
  • HyperFrames turns HTML into video. It is built for animated slides and presentations and works with Claude.
  • givemotion has an agent write HTML/CSS and render it to MP4. An open local pipeline.
  • Manim is the classic for mathematical and technical explanations, and this whole field grew out of it.

What to do with this

For a product manager or developer, the takeaway is practical. Templated videos such as release notes, weekly metrics, explainer clips and presentations can be built as code: versioned in git, fed fresh data and re-rendered in minutes. If your team has no motion designer, start with the visual tools Hera or Jitter. If you are used to the terminal, look at Remotion or Motion Canvas with an agent on top.

Some honest caveats: this is not a replacement for video models. Code will not draw a photorealistic scene or a living character. It is strong at graphics, text, diagrams and interfaces. And Motion Studio does not yet have detailed pricing or limits, so it is too early to compare it with competitors on cost. But the direction is clear: motion graphics is becoming one more task for code generation.

Story

Key Questions Answered

Why code instead of a video model?

A typical video model redraws every frame from scratch. If an intro has a typo in the headline or a wrong number on an infographic, you have to redo the whole clip, and the result comes out slightly different each time. In the code-rendered approach, the animation is a program: it has layers, points where elements appear, and timings. Want to shift a headline by half a second or change a number? You edit a parameter instead of regenerating everything.

Who else is playing in this space?

Claude Motion is the closest analogue. You describe a clip in words, Claude writes the animation as code, and you can then change the text, numbers and motion and export MP4. It does not use a video generation model. The limitation: it is still a beta for Team and Enterprise plans and is not available on the regular Pro plan. Hera is an AI motion graphics editor with editable layers. It makes ad intros, infographics, animated headlines and product videos.

What are the key details on Engines you can call from agents?

If you don't need a ready-made service, there are engines underneath, and you can call them directly or from your own agents: Remotion turns React/TypeScript into video. A versatile option for motion graphics, ads and automation. Motion Canvas uses TypeScript for vector animation. Good for explainer videos, diagrams and syncing with voiceover. HyperFrames turns HTML into video. It is built for animated slides and presentations and works with Claude. givemotion has an agent write HTML/CSS and render it to MP4.

ZK
Zhemal Khamidun — CPO at Alpina Digital, product manager of AlpinaGPT. Building AI products since 2019. Hamidun News is a pet project that grew into a production aggregator with 6,000+ readers.
✦ Unified AI Workspace · Nomad

Test Frontier LLMs & Autonomous Agents in One Workspace

Run Claude Opus 5.5, GPT-6, Gemini 3.1, and DeepSeek side-by-side in Nomad AI (nomadnet.ai) — unified model routing, deep research, and multimodal workflows under one subscription.

Try Nomad AI →Launch Nomad Agent
What do you think?
Loading comments…