If you’ve been refreshing Google’s model page waiting for this one, you’re not alone — and you’re not missing anything. Gemini 3.5 Pro was announced back in May 2026 with a lot of fanfare, and as of this writing it still hasn’t reached general availability, after slipping past three separate target windows. Here’s exactly where things stand, what’s actually confirmed, and what to use in the meantime.
What Is Gemini 3.5 Pro?
Gemini 3.5 Pro is Google’s next flagship reasoning model, announced on stage at Google I/O on May 19, 2026 alongside its sibling, Gemini 3.5 Flash. While Flash shipped that same day and is live in the Gemini app, Search’s AI Mode, and the API, Pro was held back for further development — Google’s CEO told the developer crowd to “give us until next month,” a line that reportedly drew audible groans at the time.
A Timeline of Delays
Gemini 3.5 Pro’s release has slipped repeatedly:
- May 19, 2026. Announced at I/O with a June general-availability target.
- June 30, 2026. June window closes with no public launch; Google cites quality refinements and pushes to July.
- July 17, 2026. A widely rumored date, partly driven by prediction-market speculation, also passes without a release.
- July 21, 2026. Instead of Pro, Google ships three other models — Gemini 3.6 Flash, 3.5 Flash-Lite, and a specialized 3.5 Flash Cyber for security use cases — with no mention of Pro.
- Early August 2026. Still unreleased. Google’s model pages continue listing it as “coming soon,” and some analysts now flag a real possibility the model gets folded into a future Gemini 4 release instead of shipping separately.
What’s Actually Confirmed vs. Rumored
Confirmed by Google: Gemini 3.5 Pro exists, is in limited enterprise preview through Vertex AI, and is intended to be the successor to Gemini 3.1 Pro with deeper reasoning for complex, long-horizon tasks. Everything else circulating online — a rumored 2-million-token context window, a “Deep Think” reasoning layer, and specific benchmark comparisons against GPT-5.6 and Claude — comes from leaks and speculation rather than an official model card. Treat any post claiming confirmed specs or a locked release date with real skepticism until Google publishes documentation.
Expected Pricing
Google hasn’t published official pricing for Gemini 3.5 Pro. Based on where Gemini 3.1 Pro currently sits (roughly $2–$18 per million tokens) and Google’s general pattern of pricing Pro-tier models below rival flagships like GPT-5.6 Sol, expect it to land in a similar or slightly higher range once it ships — but treat any specific number you see elsewhere as an estimate, not a confirmed figure.
What to Use Right Now
Gemini 3.5 Flash is the live, public option today and already outperforms older Pro-tier models on several agentic and coding benchmarks, making it a genuinely capable stand-in while Pro remains unreleased. Enterprise customers can request allowlist access to the limited Vertex AI preview through their Google Cloud account team, though this isn’t available to individual consumers.
Frequently Asked Questions
Is Gemini 3.5 Pro available to the public yet?
No. As of this writing, it remains in limited enterprise preview through Vertex AI only, with no confirmed general-availability date.
What’s the difference between Gemini 3.5 Flash and 3.5 Pro?
Flash is the live, public model available today in the Gemini app and API. Pro is the unreleased, more capable flagship model aimed at deeper reasoning on complex tasks, still in limited testing.
Why does Gemini 3.5 Pro keep getting delayed?
Google has cited quality refinements after early enterprise testing, and reports have described a fuller technical rebuild than originally planned, without a specific public explanation for each missed window.
What should I use while waiting for Gemini 3.5 Pro?
Gemini 3.5 Flash is the strongest publicly available Gemini model right now. For an overview of the Gemini app and how to access it, see our guide on how to use Gemini.
Final Thoughts
Gemini 3.5 Pro is real, it’s coming, and Google has been consistent about that much — but every specific date attached to it so far has turned out to be wrong. Until Google publishes an official model card and pricing, the safest move is treating it as a watchlist item rather than something to plan around. In the meantime, Gemini 3.5 Flash and other current models, covered in our roundup of new ChatGPT features, are worth evaluating instead.
