Unknown direct
See official pricing source (opens in a new tab)
Unknown
No pricing published
Image generationDocumented overviewLast checked Oct 2, 2026
Google Labs experiment that creates images from image prompts (subject, scene, style) using Gemini and Imagen 3; URL now redirects to Google Flow.
Our take: Whisk was a Google Labs image-remix experiment whose URL now leads to Google Flow, so verify availability before relying on it.
Main limitation: The standalone tool URL redirects to Google Flow, which shows only a sign-in page, and no pricing is published.
Whisk was a Google Labs image-remix experiment whose URL now leads to Google Flow, so verify availability before relying on it. Based on the vendor's documentation, not on a hands-on test.
Also a good fit for
Or if
Prices checked Oct 2, 2026 on the official US pricing page. Prices and limits change often.
Unknown
Unknown
See official pricing source (opens in a new tab)
Unknown
No pricing published
Unknown
Unknown
Whisk is a Google Labs experiment announced in December 2024 that lets people create images by prompting with images instead of long text prompts. Users drag in one image for the subject, one for the scene and one for the style, then remix them, for example into a digital plushie, enamel pin or sticker. According to Google's announcement, Gemini automatically writes a detailed caption of the input images and feeds it into Imagen 3, so the result captures the essence of the subject rather than an exact copy. It launched in the US at labs.google/whisk; that URL now redirects to Google Flow, so the standalone tool's current status is not confirmed.
You provide
You get back
Scenarios are worked out by our editors from the documentation; the actual result depends on your input and plan.
Uses images as prompts instead of long text. People who find prompt writing hard can steer the result by showing pictures instead of describing them.
Separates subject, scene and style inputs. Designers can mix a subject from one image with a scene and a style taken from other images.
Auto-captions input images with Gemini. Users do not have to write a caption for their images because Gemini describes them automatically.
Generates the final image with Imagen 3. The final image is made from the captions, so it captures the essence of your inputs rather than copying them.
Summarized from the vendor's current policies; this is not legal advice and a vendor's badge is not an independent audit.
Not stated on the pages read
Not stated on the pages read
Not stated on the pages read
Not stated on the pages read
Not stated on the pages read
Policies: blog.google/innovation-and-ai/models-and-research/google-labs/whisk/
"Better suited" is based on documented features and pricing, not on a head-to-head test.
| Alternative | When to choose it | Pricing | Limitation | |
|---|---|---|---|---|
| Google Gemini | — | — | ||
| Midjourney | — | — | Compare | |
| Adobe Firefly | — | — | Compare | |
| Leonardo.Ai | — | — | Compare |
No user reviews yet.
Used Whisk (Google Labs)? Be the first to share your experience.
A Google Labs experiment that builds images from subject, scene and style images using Gemini and Imagen 3. source (opens in a new tab)
The Whisk URL now redirects to Google Flow. source (opens in a new tab)
Yes, sign-in is required. source (opens in a new tab)
Unknown; no pricing was published. source (opens in a new tab)
Not verified: Current status; Pricing; Watermark rules.
Every fact about Whisk (Google Labs) above comes from the 2 official pages listed here, last compared on Oct 2, 2026; we have not run Whisk (Google Labs) ourselves, so it has no editorial score. How we check and re-check pages.
Article updated Oct 2, 2026; prices checked Oct 2, 2026. Outbound links to vendors are plain links; we do not earn commissions. How we review · Report an error
compare>