Google's promised Gemini 3.5 Pro is still missing after June pledge
Sundar Pichai said Gemini 3.5 Pro would arrive in June, but the model remains unreleased as rivals roll out fresh AI systems.
By Georgia Hale · Staff Writer
3 min read
Google’s next top-tier Gemini model was supposed to arrive in June. As of July 17, Gemini 3.5 Pro still has not appeared.
At Google I/O 2026 in May, the company rolled out Gemini 3.5 Flash, a smaller model aimed at everyday use. Google CEO Sundar Pichai also told reporters before the event that the more powerful Pro version was in internal use, showing gains, and set for release the following month.
“We are also excited for 3.5 Pro,” Pichai said at the media briefing. “We are using it internally. It's showing great improvements. We are still testing and refining it, and it will roll out to everyone next month.”
That June window has now come and gone, and Google has not announced a public launch date.
Google says testing is still underway
Asked by Mashable about the timeline, Google gave the same statement it provided to Bloomberg after that outlet reported on the delay.
“We’re shipping quickly across a wide range of models while keeping them highly cost-effective for customers. We’re currently testing 3.5 Pro, an upgraded Flash model, and other models with partners, and we’re productively engaged with the U.S. government on model testing and broader frameworks.”
Bloomberg reporters Julia Love and Davey Alba reported on July 16 that the delay has frustrated some Google engineers, AI researchers and managers. According to Bloomberg, those staffers are worried Google could lose ground to rivals including Anthropic and OpenAI.
The report pointed to two factors behind the holdup. One is the scale of Google itself: Gemini is tied into many products, and the company’s size can slow releases compared with smaller AI firms. The second, according to Bloomberg, is concern among Google leaders that Gemini 3.5 Pro may not match the newest rival models.
Rivals are not waiting around
Since Google I/O, several competitors have put fresh AI models into the market or announced them.
- Anthropic announced Claude Mythos Preview, which it described as its most advanced model, with cybersecurity abilities strong enough that it would initially be limited to trusted partners.
- Anthropic later released a version of Claude Mythos called Fable 5 on June 9.
- OpenAI announced GPT-5.6 Sol on July 9, describing it as a next-generation model with advanced cybersecurity coding abilities.
- Chinese AI lab Moonshot released Kimi K3 this week, an open-source model with 2.8 trillion parameters. Early testers cited by Artificial Analysis said its abilities were similar to Fable 5 and GPT-5.6 Sol at a lower cost.
Bloomberg also reported that Meta has released a model that beats Google Gemini. The delayed Pro launch leaves Google without a new frontier model while rivals keep raising the benchmark.
Google still has built-in strengths in the AI race, including access to enormous amounts of data and the ability to put AI tools in front of Android users worldwide. For now, though, Gemini 3.5 Pro remains in testing, and Google has not said when the public will get it.
This story draws on original reporting from Mashable.