innFactory AI Consulting, based in Rosenheim, Germany, supports enterprises across the DACH region (Germany, Austria, Switzerland) and Europe with GDPR-compliant deployment of Google Gemini. On 21 July 2026, Google released three new models: Gemini 3.6 Flash (GA) as the new workhorse model with better coding and 17 percent fewer output tokens than 3.5 Flash, Gemini 3.5 Flash-Lite (GA) as the fastest model in the 3.5 series at 350 output tokens per second, and Gemini 3.5 Flash Cyber as a specialised security model in a limited-access pilot for governments and vetted partners. Gemini 3.5 Pro, by contrast, keeps slipping: Google is still testing the orchestrator model with partners and no longer names a GA date. In parallel, DeepMind confirmed that the pre-training run for Gemini 4 is already underway.
EU availability (as of 24 July 2026): For GDPR-compliant workloads, Gemini 3.5 Flash remains the reference – it is available via the EU multi-region endpoint on Vertex AI / Gemini Enterprise Agent Platform; requests are routed exclusively inside the EU geography and covered by the Vertex AI Data Processing Addendum. For the new Gemini 3.6 Flash, EU multi-region availability is not yet confirmed – though for 3.5 Flash the EU endpoint followed just two days after GA; we are monitoring continuously. A single-region endpoint in
europe-west3(Frankfurt) oreurope-west4(Netherlands) is still missing for the 3.x family; for strict single-region pinning, Gemini 2.5 Pro and Gemini 2.0 Flash ineurope-west4remain the path.
Key Strengths
Gemini 3.6 Flash – The New Workhorse (GA since 21 July 2026)
Gemini 3.6 Flash (gemini-3.6-flash) improves coding, knowledge work, and multimodal processing over 3.5 Flash while being cheaper: $1.50/1M input, $7.50/1M output. On the DeepSWE coding benchmark the model jumps from 37 to 49 percent, and it reaches 83 percent on OSWorld-Verified (computer use). At the same time it produces 17 percent fewer output tokens than 3.5 Flash – Google directly addressing developer feedback on verbosity, which lowers effective per-task cost further. The knowledge cutoff advances to March 2026; the context window stays at 1 million input tokens (65,536 output). The model is available via the Gemini API (AI Studio, Android Studio), Antigravity, the Gemini Enterprise Agent Platform, and the Gemini app. Migration note: the sampling parameters temperature, top_p, and top_k are deprecated on the new models – review existing integrations.
Gemini 3.5 Flash-Lite – Speed for High-Volume Workloads (GA since 21 July 2026)
Gemini 3.5 Flash-Lite (gemini-3.5-flash-lite) is the fastest model in the 3.5 series: 350 output tokens per second as measured by Artificial Analysis, priced at $0.30/1M input and $2.50/1M output. That makes it a fit for high-volume automation, classification, and latency-critical applications – for example as a sub-agent in multi-agent pipelines. Google is additionally rolling the model out in Google Search.
Gemini 3.5 Flash Cyber – Security Model in a Pilot Program
With Gemini 3.5 Flash Cyber, Google announced its first dedicated cybersecurity model, paired with the CodeMender security agent. It remains exclusive for now: access is limited to governments and vetted partners via a limited-access pilot program. The model is therefore not currently relevant for enterprises, but it shows Google’s course of rolling out security-critical capabilities in a controlled way – a pattern we already know from the government clearance processes at OpenAI GPT-5.6 and Anthropic Claude Fable 5.
Gemini 3.5 Flash – Agentic-first
With Gemini 3.5 Flash, Google positions its fastest tier above its own Pro flagship for the first time: 3.5 Flash outperforms Gemini 3.1 Pro on Terminal-Bench 2.1 (coding), GDPval-AA Elo (real-world agentic) and MCP Atlas (tool use). The model is built for multi-hour autonomous workflows – coding pipelines, research projects, even building entire systems – pausing for human input at decision points. An optimised variant is reported to reach up to 12× the speed of other frontier models at equivalent quality.
Gemini 3.5 Pro – Orchestrator (Further Delayed)
Gemini 3.5 Pro remains in partner testing: the GA originally expected for June 2026 has slipped, and in the July 21 release blog post Google only states the model will be “broadly available as soon as it’s ready”. The model is designed as the orchestrator/planner that drives Gemini 3.5/3.6 Flash instances as sub-agents. Google has announced a 2M token context window and a Deep Think reasoning mode for Pro, explicitly positioning the family for multi-agent architectures rather than classic chatbots.
Gemini Spark – Agentic 24/7 Assistant
At Google I/O 2026, Google introduced Gemini Spark, an agentic personal assistant built on Gemini 3.5 and the Antigravity platform. Spark runs continuously on Google’s cloud, can be addressed directly via a dedicated Gmail address, and executes background tasks across Chrome and Google Workspace. Spark supports the Model Context Protocol (MCP) for external integrations and is rolling out first to Google AI Ultra subscribers.
Industry-Leading Context Window
Gemini 3.1 Pro processes up to 2 million tokens in a single context. This enables analysis of extensive contract portfolios, technical documentation, or entire codebases in a single request. The Flash models offer a 1 million token context window at significantly lower cost.
Native Multimodal Processing
The Gemini 3.x family natively processes text, images, audio, video, and PDF documents. This enables use cases such as automated video analysis, document extraction from scanned PDFs, or meeting evaluation combining audio and visual material.
Google Ecosystem
- Vertex AI: Enterprise deployment with SLA (EU hosting only for older models)
- Antigravity 2.0: Agentic development platform for Gemini 3.5 and 3.6
- Gemini Enterprise: Enterprise frontend with access to 3.6 Flash and 3.5 Flash
- Google Workspace: Integration with Docs, Sheets, Gmail, and more
- Search Grounding: Access to current web information
- Google AI Studio: Rapid prototyping and API access
Specialised Variants
- Gemini 3.1 Flash Thinking: Extended reasoning for complex STEM tasks with a transparent thinking process
- Gemini 3.1 Pro Deep Research: Specialised for multi-step research and long analysis tasks
EU Availability
As of 24 July 2026: Gemini 3.5 Flash is GA on the EU multi-region endpoint on Vertex AI / Gemini Enterprise Agent Platform. Requests are routed exclusively inside the EU geography and the model is covered by the Vertex AI Data Processing Addendum. For the new models Gemini 3.6 Flash and Gemini 3.5 Flash-Lite (GA since 21 July), EU multi-region availability is not yet confirmed. Gemini 3.5 Pro remains in partner testing without a GA date.
What works from EU projects today
- Gemini 3.5 Flash via the EU multi-region endpoint — GDPR-compliant routing inside the EU geography, GA on Vertex AI / Gemini Enterprise Agent Platform.
- Gemini 3.6 Flash and 3.5 Flash-Lite via the global endpoint — available for workloads that don’t require regional binding.
- Gemini 2.5 Pro and Gemini 2.0 Flash as single-region regional endpoints in
europe-west4(Netherlands), partial coverage ineurope-west3(Frankfurt) — for customers that need strict pinning to a specific EU country.
What is still missing
- Confirmed EU multi-region availability for Gemini 3.6 Flash and Gemini 3.5 Flash-Lite — for 3.5 Flash the EU endpoint arrived two days after GA (19 → 21 May 2026); we expect a similar pattern and are monitoring continuously
- Single-region endpoints for the 3.x family in
europe-west3(Frankfurt) oreurope-west4(Netherlands) - Gemini 3.5 Pro (partner testing, GA date open) — EU availability to be confirmed at GA launch
We recommend monitoring the Google Cloud regional availability documentation and the Gemini Enterprise Agent Platform release notes regularly.
Integration with CompanyGPT
Gemini models are integrated in CompanyGPT. Gemini 3.5 Flash is wired in via the EU multi-region endpoint for GDPR-compliant frontier reasoning and agentic coding workloads. Gemini 2.5 Pro and Gemini 2.0 Flash remain available as single-region endpoints in europe-west4 for customers that require strict country-level data residency. As soon as Google ships single-region 3.5 endpoints in Frankfurt or the Netherlands we will add them as default options.
Our Recommendation
Since 21 July 2026, Gemini 3.6 Flash is Google’s most capable GA model – but for GDPR-compliant EU workloads, Gemini 3.5 Flash remains the recommendation until the EU endpoint is confirmed:
- Frontier reasoning and agentic workflows in EU: Gemini 3.5 Flash via the EU multi-region endpoint – switch to 3.6 Flash as soon as Google confirms the EU endpoint (we will update this page)
- Workloads without EU residency requirements: Gemini 3.6 Flash via the global endpoint – better coding at a lower price than 3.5 Flash
- High-volume and latency-critical: Gemini 3.5 Flash-Lite ($0.30/1M input, 350 tokens/s) via the global endpoint
- Strict single-country EU residency: Gemini 2.5 Pro in
europe-west4 - Cross-cloud alternative: Anthropic Claude Opus 4.8 or Claude Sonnet 4.6 via AWS Bedrock (in-region Ireland / Stockholm / Frankfurt) — useful for hybrid Multi-Cloud strategies or when the workload is not bound to Google Cloud
We are closely tracking the EU rollout of Gemini 3.6 Flash and 3.5 Flash-Lite and will update this page once Google ships the endpoints.