The short version: Google Gemini 3.5 Pro was announced at Google I/O on 19 May with a firm intention to launch "next month" — which makes June the window. Prediction markets put the probability of a before-June-30 release at 50–55%. The model targets a 2-million-token context window (enough to process a library of business documents in a single session), a Deep Think reasoning mode for complex analysis, and performance that early tests suggest is four times faster than comparable frontier models. For small businesses on Google Workspace, this is a meaningful capability upgrade arriving at no additional cost.
What Gemini 3.5 Pro offers
Current AI tools typically handle context windows of 100,000 to 200,000 tokens — enough for a long document or a lengthy conversation. Gemini 3.5 Pro's 2-million-token window is large enough to feed in an entire year of business emails, a complete contract library, a full customer correspondence archive, or hundreds of meeting transcripts in a single session. This does not just improve AI outputs — it enables new categories of task that were not previously feasible for small business operators.
The other major addition is Deep Think — a reasoning mode that produces longer, more structured analytical outputs for complex questions. Where the current Gemini handles direct queries well, Deep Think is designed for tasks like: analysing a set of contracts for risk, reviewing months of sales data for patterns, or preparing a comprehensive business report from multiple source documents. These are tasks that previously required hours of human analysis — or a consultant's invoice.
Speed: four times faster than comparable models
Performance testing of Gemini 3.5 Flash (the faster sibling, already released) has achieved 284 tokens per second — approximately four times faster than competing frontier models at similar intelligence levels. Gemini 3.5 Pro is expected to be somewhat slower due to its reasoning capabilities, but still meaningfully faster than GPT-5.4 at comparable output quality.
For business operators, speed in AI tools is not a benchmark curiosity. When you are waiting for a draft email, a document summary, or a customer response suggestion, the difference between three seconds and fifteen seconds is the difference between a tool you actually reach for and one you put down. Speed is a usability factor, not a technical one.
Pricing and access for UK businesses
Based on Google's track record with Gemini releases, the model will be available in Gemini Advanced (£18.99/month) and Google One AI Premium (which includes Workspace integration) before it becomes available via API. For UK businesses already on Google Workspace Business Standard or Business Plus, access via the Workspace AI features is the expected path. API pricing is anticipated at approximately £12 per million input tokens and £48 per million output tokens — roughly ten times the cost of Gemini 3.5 Flash.
What to prepare now
Most small businesses will not need to do anything technical to access Gemini 3.5 Pro — it will roll out through the same Google apps they already use. The preparation that matters is knowing how to use it when it arrives:
Four tasks to test on day one
1. Document analysis at scale: Upload a year of contracts, invoices, or correspondence and ask Gemini to summarise patterns, flag anomalies, or extract specific information. The 2M context window makes this possible for the first time at this level of depth.
2. Meeting transcript processing: If you record and transcribe meetings, test feeding multiple transcripts in one session and asking for action items, decisions, and follow-up tracking across all of them.
3. Deep Think for business planning: Give it a complex analytical question — "What are the main risks in my current pricing structure based on these customer conversations?" — and compare the output quality to current Gemini Pro and ChatGPT.
4. Customer communication drafting: Draft a set of customer responses or proposals in your voice. With a larger context window, Gemini can absorb more of your existing communications before drafting — producing outputs that are closer to your style without manual editing.
The practical opportunity here for UK small businesses is access to a level of AI capability — specifically, the ability to process large document sets with structured reasoning — that was previously only accessible to larger organisations with technical teams or dedicated AI infrastructure. When Gemini 3.5 Pro arrives in Google Workspace, it arrives at the price of a seat you already own.