Making a short explainer? Renderly now generates its illustrations too.

Renderly is a template-based video service with an editor and an API. The new release adds image generation inside the render request, a separate image-generation endpoint, and an image tool for connected AI assistants. The announcement says the assistant tool is available on API-key connections. This is not a new one-button video app for everyone.

A shared style instruction can apply across the pictures in a video; individual pictures can override it. The service also reads the shape of the image slot so the generated picture has the matching proportions. Those are sensible details for an explainer that should look like one piece rather than a pile of unrelated images.

Renderly demonstrates the release with a six-image Kyoto guide. It reports a 34-second finished video costing 3.7 credits, including image generation and rendering. That is the maker's own example, not an independent speed or quality test. We have reviewed the announcement and documentation, not run the service.

Imagine a small pottery shop explaining how it packs a mug for delivery. A drawn sequence can show cushioning, a closed box and the parcel's journey without requiring a new photo shoot for every step. A real photograph should still show the mug's actual glaze and shape. That is an example of how I would use the feature, not a customer result reported by Renderly.

The release allows generated images and existing media to be mixed in one render. Use that distinction deliberately: actual objects, venues and finished work can stay as supplied photographs, while illustrations cover the ideas that do not need photographic proof. Ask for a plainly illustrated look when a realistic invented scene could be mistaken for something that happened.

This applies to the script as well. A generated picture of careful packing does not establish that a shop uses that method. Start from an explanation the business can stand behind, then choose pictures that help someone understand it.

Renderly says it stores generated images and reuses an image without another image-generation charge when the prompt, style, tier, size, aspect ratio and seed are identical. Change the text of the video while keeping those image settings unchanged, and the pictures can stay the same. A different seed requests a fresh version.

That matters after someone has approved the visuals. Correcting a sentence should not send you back through a gallery of slightly different mugs. It also makes a small correction less expensive than starting the whole piece again, although the video render itself is still billed.

The pricing documentation says rendering normally costs one credit per started minute, with some templates carrying a higher multiplier. Image generation adds its own charges. The announcement lists 0.5 credits for a quality-tier 1K image and 0.2 for a fast-tier 1K image. Credits are not dollars; check the account's actual purchase price before budgeting a batch.

If you already use Renderly or have someone who can connect it, start with an explanation you keep sending anyway. Make one short version using approved words, your own necessary photographs and a few clearly illustrative images. Watch it on the phone where the intended viewer will see it.

Then change a sentence. Check whether the approved pictures stay put, the text is still readable and the new export says what you meant. That second version will tell you more about living with the tool than a spectacular first generation.

If setting up the connection costs more effort than making your occasional video, leave it for now. A simple page with a good photo may already answer the question. The worthwhile result is an explanation you no longer have to rebuild, not another publishing schedule to maintain.