What to know
- Gemini’s image editor is adding object discovery and a long-press gesture for placing targeted edit points.
- The update extends the existing dot-based editor with an explicit “Add an edit” mode and clearer point controls.
- We found the changes in code from the unreleased Google app v17.66.11.sa.arm64 and have not surfaced them in the app yet.
Gemini’s broader image-editing capability, including draw-to-edit tools, is already public. The latest Google app build now points to a more precise way to direct those edits: touch and hold at a specific place in an image to add an edit point.
We diffed Google app v17.66.11.sa.arm64 against the earlier build and found new strings covering point placement, object discovery, accessibility guidance, and the editor’s failure state.
Long-pressing adds a targeted edit point
The clearest addition is a new “Add an edit” mode. Updated guidance tells you that you can still tap an existing dot to edit it, or touch and hold to place another one. Separate accessibility text describes that action as adding an edit “at this point,” reinforcing that the gesture targets a particular part of the image rather than starting a general edit.
The new strings, seen below, point to that interaction and its accompanying object-discovery step:
The “Couldn’t find objects” message shows that Gemini can look for selectable objects while preparing this editing interface. It also gives the flow a defined failure state when that discovery step does not return an object.
This expands an editor already present in the code
This is an update to an existing Gemini image-editing feature, not the first sign of the feature itself. Its earlier code appeared in Google app v17.63.17.ve.arm on October 2, 2026, with controls for removing a selected object and tapping existing dots to edit them.
Google app v17.66.11.sa.arm64 retains that point-based foundation while adding the long-press creation path. A separate new “Saving edit” label also accounts for the period when an edit is being stored.
Together, the strings describe a more deliberate editing flow: Gemini identifies objects, presents editable points, lets you add a point exactly where another change is needed, and then shows a saving state for the resulting edit.
The updated point-editing interaction is not available to users yet, and we’ll watch future Google app builds for signs of its rollout.
Related on NerdsChalk
Leave a Reply