What happened
Multiple outlets reported on June 30 that Gemini 3.5 Pro had missed its June general-availability deadline and now targets July 2026, with the delay attributed to feedback about token consumption in extended agentic tasks. Google has not officially confirmed a revised date. Expected specs for 3.5 Pro include a 2-million-token context window and a Deep Think reasoning mode.
To keep its lineup moving, Google released two new image-generation models the same day: Gemini 3.1 Flash Image, priced at $0.50 and $3.00, and Gemini 3 Pro Image, priced at $2.00 and $12.00, both in Google AI Studio and the Gemini API. It is the second consecutive missed I/O commitment, coming amid senior Gemini researcher departures.
Why it matters
Token consumption in long agentic tasks is a real cost and reliability problem, so a delay to fix it is defensible. But a second consecutive missed commitment raises questions about Google's release discipline.
Shipping image models to fill the gap keeps the product lineup active while the flagship slips. The researcher departures add a note of caution about momentum inside the Gemini group.
MintedBrain take
A delay tied to token efficiency is preferable to shipping an expensive-to-run flagship. Still, the pattern of missed dates is worth tracking; plan around the models that are actually available, not the ones on the roadmap.
Discussion
Sign in to comment. Your account must be at least 1 day old.