- Google Japan has released the July 2026 edition of “Gemini Drop,” introducing new Google AI “Gemini”-related features announced this month (July 2026).
- The “Gemini Drop” released for Japan / Japanese consists of four features: the “Avatar” generation feature update, the global rollout of “Gemini Spark,” voice control for the macOS version of the “Gemini” app, and “Gemini 3.6 Flash” and “Gemini 3.5 Flash-Lite.”
- However, the features currently rolled out in the Japan / Japanese environment are only two: the global rollout of “Gemini Spark” and “Gemini 3.6 Flash” / “Gemini 3.5 Flash-Lite.”
On Friday, July 31, 2026, Google’s Japanese subsidiary, Google Japan, released the July 2026 edition of “Gemini Drop,” which introduces new Google AI “Gemini”-related features announced this month (July 2026).
The lineup for the July 2026 edition of “Gemini Drops” released globally by Google consists of a total of six features: voice control for the macOS version of the “Gemini” app, the global rollout of “Gemini Spark,” “Gemini 3.6 Flash” and “Gemini 3.5 Flash-Lite,” the “Avatar” generation feature update, expanded “App Integration” for “Gemini,” and personalized image generation via “Personal Intelligence.” Among these, the “Gemini Drop” released for Japan / Japanese includes four features: the “Avatar” generation feature update, the global rollout of “Gemini Spark,” voice control for the macOS version of the “Gemini” app, and “Gemini 3.6 Flash” and “Gemini 3.5 Flash-Lite.”
However, the features currently rolled out in the Japan / Japanese environment are just two: the global rollout of “Gemini Spark” and “Gemini 3.6 Flash” / “Gemini 3.5 Flash-Lite.” The “Avatar” generation feature update and voice control for the macOS version of the “Gemini” app are expected to be rolled out sequentially in the Japan / Japanese environment.
jetstream.blog“Avatar” Generation Feature Update
The “Avatar” generation feature available in the Google AI “Gemini” app has been updated, eliminating the need to re-upload selfie videos or recorded audio when generating an “Avatar” again.
That said, the “Avatar” generation feature in the “Gemini” app remains rolled out for English environments.
jetstream.blogGlobal Rollout of “Gemini Spark”
The personal AI agent “Gemini Spark” has expanded its rollout globally, including Japan.
Along with the expanded rollout of “Gemini Spark,” integration with Google Workspace services has also been enhanced. This includes editing files in Google Sheets and Google Slides, reading out “comments” in Google Sheets and Google Slides, editing shared Google Docs, Google Sheets, and Google Slides files, adding images to Google Docs and Google Slides files, and editing and revising Google Docs using the “Canvas” tool. In addition, the speed of “Gemini Spark” itself has been increased by more than 50%.
However, “Gemini Spark” is not available in the European Economic Area (EEA), the UK, Switzerland, and Nigeria.
Voice Control for the macOS “Gemini” App
A new feature has been added to the macOS version of the “Gemini” app, allowing voice control across any window by simply holding down the Fn key.
By default, it operates as a voice dictation feature that automatically removes fillers like “um” or speech errors from what you say, inserting it directly as formatted text at the cursor position. When the “Gemini” reasoning feature is turned on in settings, even more advanced tasks become possible, such as having it summarize highlighted content on the screen using only your voice, rewriting text with a different tone, or giving voice instructions to generate and edit images.
This new feature for the macOS “Gemini” app is initially rolling out globally for English environments. Support for additional languages is planned to expand sequentially in the future.
jetstream.blog“Gemini 3.6 Flash” and “Gemini 3.5 Flash-Lite”
“Gemini 3.6 Flash” and “Gemini 3.5 Flash-Lite” have been announced as new AI models for Google AI “Gemini.”
“Gemini 3.6 Flash” is a flagship AI model with improved processing efficiency and reduced operational costs compared to the previous AI model, “Gemini 3.5 Flash.” “Gemini 3.5 Flash-Lite” is the fastest and most lightweight AI model in the “Gemini 3.5” family.
Source: Google (https://gemini.google/gemini-drops/)