At Google I/O 2026, Google announced two new models: Gemini Omni and Gemini 3.5 Flash. Google describes Gemini Omni as a model that "can create anything from any input, starting with video", marking a leap forward in world understanding, multimodality, and editing.
Gemini 3.5 Flash is described as the first in Google's latest family of models "combining frontier intelligence with action". Alongside both launches, Google also restructured its consumer AI subscription pricing, reshaping how access to these models actually works. Let us discuss in detail.
Gemini Omni's Core Capability
Gemini Omni is Google DeepMind's new model built to generate and edit video from any combination of image, audio, video, and text inputs. Google states that with Omni, users can combine images, audio, video, and text as input and generate high-quality videos grounded in Gemini's real-world knowledge.
What Makes Gemini Omni Different
Google highlights conversational editing as the defining capability of Omni. Video can be edited through natural language instructions, with characters remaining consistent and scene context carried forward across edits, so each instruction builds on the last rather than starting from a blank prompt.
A new YouTube Shorts Remix feature also lets users select an eligible short and prompt changes, such as adding themselves or a visual reference, to generate a new version of it.
Google Gemini Omni Pricing and Access
Gemini Omni Flash is rolling out to all Google AI Plus, Pro, and Ultra subscribers globally through the Gemini app and Google Flow and is available at no cost on YouTube Shorts and the YouTube Create app.
Gemini 3.5 Flash's Core Capability
Gemini 3.5 Flash is the first release in Google's new Gemini 3.5 family. Google states the model delivers frontier performance for agents and coding, excelling at complex long-horizon tasks that deliver real-world utility while running at the speeds expected from the Flash series.
Performance and Benchmarks for Gemini 3.5 Flash
Google states that compared to Gemini 3.1 Pro, Gemini 3.5 Flash performs better across almost all benchmarks, with a notable jump on GDPVal, which Google describes as capturing real-world, economically valuable tasks.
On output tokens per second, Google states the model runs four times faster than other frontier models. Gemini 3.5 Pro is already in use internally at Google and is expected to roll out publicly the month following the Flash launch.
Gemini 3.5 Flash Pricing Structure
Gemini 3.5 Flash is generally available through Google Antigravity, the Gemini API in Google AI Studio, Android Studio, the Gemini Enterprise Agent Platform, and Gemini Enterprise, and is also available to everyone in the Gemini app and in AI Mode in Search, where it is now the default model globally.
Google AI Pro and Ultra subscribers also receive Google Cloud credits, $10/month for Pro and $100/month for Ultra, intended to support moving AI projects from prototype to production.
Also read USAII®'s latest insight on OpenAI Swaps ChatGPT's Core Model: A Closer Look at GPT-5.5 Instant to see how a comparable default-model shift is playing out at OpenAI, where GPT-5.5 Instant replaced GPT-5.3 Instant in May 2026 with measurable gains in factual accuracy and reasoning.
Gemini Omni Vs Gemini 3.5 Flash: What Each Is Built For
The two models are not competitors within Google's own lineup. Omni is built for multimodal AI content creation, while 3.5 Flash is built to power agents that plan, use tools, and execute long-horizon tasks.
What This Means for AI Professionals
The pace of model releases in 2026 is reshaping what AI fluency means. Prompting one model well is no longer enough; professionals now need to understand how different model families reason, where their capabilities diverge, and which model fits which task. Multimodal generation, agentic workflows, and long-horizon reasoning are fast becoming baseline expectations, not specialized AI skills.
For professionals looking to build that competency formally, the Certified Artificial Intelligence Engineer (CAIE™) by USAII® is a professionally recognized generative AI certification that covers large language model architecture, tools and frameworks for LLMs, prompt engineering, and retrieval-augmented generation, giving professionals the foundational AI engineering skills needed to work with evolving model families as new releases like Gemini Omni and Gemini 3.5 Flash reach the market.
Way Forward
Gemini Omni and Gemini 3.5 Flash represent two different directions in Google's AI strategy, one toward unified multimodal content creation and the other toward faster, more capable agentic execution, both now backed by a restructured pricing model designed to bring more users into paid tiers.
The AI professionals who stay ahead are the ones who build a working understanding of what each model is actually built for and what it costs to access, rather than treating every new release as interchangeable with the last.
FAQs
Is Gemini 3.5 Flash available to use right now?
Yes, it is generally available through Antigravity, the Gemini API, AI Studio, Gemini Enterprise, and the Gemini app, including as the new default in AI Mode in Search.
Can developers build agentic applications with Gemini 3.5 Flash?
Yes, Google introduced Managed Agents within the Gemini API alongside Gemini 3.5 Flash, allowing developers to deploy agent workflows directly through the API.
What is Gemini Enterprise, and how does it relate to Gemini 3.5 Flash?
Gemini Enterprise is one of the platforms where Gemini 3.5 Flash is generally available, alongside the Gemini Enterprise Agent Platform, Google Antigravity, and the Gemini API.
