Explainer video guide
Open source AI video generator for explainers, no GPU needed.
For founders, teachers and product teams who want open models without a GPU rig. Paste a script, get scenes, voice and captions.
Shot list: how it works
Four steps from script to finished explainer.
Treat your script as a numbered shot list. Each line becomes one scene, and each scene gets the look you picked.
Pick an open-weight model
Choose the model that suits the look of your explainer: 2D flat, 3D or whiteboard. The style carries through every scene.
Paste your script
Drop in the script. Frameo splits it into scenes, one line per shot, so you can read the whole film before you generate it.
Generate and swap shots
Generate each scene. If one shot misses, swap it and leave the rest of the explainer as it is.
Add voice and captions
Add a voiceover and captions on the same canvas. Then export in 16:9 for the web or 9:16 for social.
Self-hosted route
Running an open source model yourself looks like this.
One shot, the long way. Then the same shot from a prompt.
> git clone open-video-model
> pip install -r requirements.txt
> python download_weights.py
fetching model weights
> python render.py --shot 01 --seconds 5 --prompt 'flat 2D phone, app opens'
waiting for GPU
SHOT 01 written: shot_01.mp4
frameo: SHOT 01 from the same prompt, ready on your canvas
Self-hosted vs Frameo
Self-hosting versus open models in Frameo.
Same explainer, two routes. Here is what each one asks of you, row by row.
| Compared | Self-hosting an open-source model | Open models in Frameo |
|---|---|---|
| Hardware needed | A capable GPU you buy or rent. | A browser. |
| Setup | Repo, dependencies and model weights. | None. Pick a model. |
| Scenes from a script | You write your own pipeline. | Paste the script. Scenes split for you. |
| Voiceover and captions | Separate tools, stitched by hand. | On the same canvas. |
| Consistent style across scenes | Manual prompts and seeds. | One look carried through every scene. |
| Export formats | Whatever your pipeline writes. | 16:9 or 9:16. |
| Cost to start | Hardware or rented GPU time first. | Credits pay for generation, not seats. The exact cost of a render is shown before you start it. |
Explainer video prompts
Eight explainer prompts ready to make.
Pick a topic, copy the prompt or start creating, then swap in your own script and details.
Open source AI video questions
No. Frameo is an AI video production platform: it runs open-weight models alongside other frontier models on one canvas, so you skip the setup and the GPU.
You can use the open-weight models offered in Frameo, picked per scene for a 2D flat, 3D or whiteboard look. Every frontier model sits on the same canvas, so you can mix them.
Yes. Paste your script and Frameo splits it into scenes. Bring your own voiceover or generate one on the same canvas.
You retain ownership of what you make. The Terms of Use set out the full position.
Have a script? Turn it into an explainer today.
Paste it in, pick a look, and Frameo splits it into scenes with voiceover and captions on one canvas.
Start creating