Google Unveils New Gemini Models: What Happened to 3.5 Pro?

TL;DR

• Google launched Gemini 3.6 Flash and 3.5 Flash-Lite today to handle everyday and high-throughput tasks, offering improved speed and lower costs than previous versions.
• The anticipated Gemini 3.5 Pro remains delayed for a third time, with Google stating it is still "testing with partners" and will launch "as soon as it's ready" rather than on a fixed date.
• A new specialized model, Gemini 3.5 Flash Cyber, was introduced for security vulnerability detection and patching, currently available only to governments and trusted partners in a limited pilot.

The New Gemini Lineup: What Actually Arrived

Google has officially expanded its Gemini family with the release of Gemini 3.6 Flash and Gemini 3.5 Flash-Lite, marking a strategic pivot toward efficiency and specialized utility while the flagship Gemini 3.5 Pro remains in the wings. Alongside these general releases, the company unveiled Gemini 3.5 Flash Cyber, a model dedicated exclusively to identifying and fixing security vulnerabilities, though it is currently restricted to a limited-access pilot for governments and trusted partners.

Why the 3.5 Pro Is Missing

The absence of the highly anticipated Gemini 3.5 Pro is not due to a cancellation but rather a significant delay. Reports indicate that the model has been postponed for the third time since its initial June 30th launch date, with internal testing suggesting the model is simply "not ready yet." Google's official stance confirms that 3.5 Pro is currently "testing with partners" and will receive broad availability only "as soon as it's ready," rather than adhering to a specific calendar deadline.

This delay appears to have prompted Google to release the 3.6 Flash variant as an interim solution to maintain momentum in the AI market. While the 3.5 Pro was expected to be the strongest agentic and coding model yet, the 3.6 Flash has already been deployed as the default model for the Gemini app and AI Mode in Search globally.

Breakdown of the New Models

The three new releases target distinct use cases, balancing performance with cost and latency.

Gemini 3.6 Flash
This model serves as the upgraded standard for frontier performance in agents and coding. It is designed to solve complex real-world problems at speed, offering frontier-level understanding across text, audio, images, code, and video. Key improvements include consuming 17% fewer output tokens compared to the previous 3.5 Flash and taking fewer reasoning steps to accomplish multi-step workflows. Its pricing is set at $1.50 per 1M input tokens and $7.50 per 1M output tokens, a reduction from the previous $9/1M output cost.

Gemini 3.5 Flash-Lite
Targeting high-throughput and low-latency tasks, this model is optimized for agentic search and document processing. Google claims it offers "significantly better quality" than the 3.1 Flash-Lite from March. It is positioned as the most cost-effective option in the new lineup, priced at $0.30 per 1M input tokens and $2.50 per 1M output tokens.

Gemini 3.5 Flash Cyber
This specialized model focuses on the "agentic era" of security, designed to detect, validate, and patch code security issues at scale. It leverages the efficiency of the Flash foundation to provide security analysis at a lower price per token than larger models. Google's CodeMender tool utilizes multiple 3.5 Flash Cyber agents to automate these tasks.

Availability and Developer Access

Both Gemini 3.6 Flash and 3.5 Flash-Lite are available immediately to users worldwide via the Gemini app. The Flash-Lite model is also integrated into Google Search AI Mode. For developers, access is provided through Google Antigravity, the Gemini API in Google AI Studio, and Android Studio.

The Gemini 3.5 Flash Cyber model has a restricted rollout. It is initially accessible only to governments and trusted partners as part of a limited-access pilot program. Users can access the 3.6 Flash model in the app by selecting "3.6 Flash" from the model drop-down menu.

Strategic Implications for Google's AI Roadmap

This release strategy suggests Google is prioritizing agentic capabilities and cost efficiency over rushing a potentially unstable flagship model. By deploying 3.6 Flash as the new default, Google ensures its users have immediate access to high-performance agents without waiting for the Pro model. The company also signaled future ambitions, revealing that it has already begun its "most ambitious pre-training run yet" for Gemini 4.

For developers, the shift means a more granular toolset: 3.6 Flash for heavy coding and reasoning, Flash-Lite for rapid, low-cost tasks, and Flash Cyber for security automation. The delay of 3.5 Pro, however, leaves a gap in the "strongest agentic" tier that the 3.6 Flash is currently filling, though the Pro model is expected to eventually outperform it on benchmarks like Terminal-Bench 2.1 and CharXiv Reasoning once it is released.

AndroGuider Team

Articles written by the AndroGuider team. We try to make them thorough and informational while being easy to read.