New AI Models Gemini 3.6 Flash and 3.5 Flash-Lite

Google has introduced new models in the Gemini family — Gemini 3.6 Flash and Gemini 3.5 Flash-Lite. Both models are designed for fast, cost-effective, and scalable AI scenarios: from code writing and document analysis to workflow automation and building AI agents.

Now you can test the new Gemini models at vega.chat — an aggregator of top AI models accessible from Russia. No need to switch between different services, deal with foreign restrictions, or pay for multiple subscriptions. Select the desired model in the vega.chat interface and compare its responses with other modern AI solutions.

Gemini 3.6 Flash: More Capabilities with Fewer Tokens

Gemini 3.6 Flash is the primary workhorse model in Google's new lineup. It targets tasks where response quality, execution speed, and token efficiency are all critical.

According to Google, Gemini 3.6 Flash:

• Shows higher performance in coding and intellectual tasks;


• Handles multi-step tasks more effectively;


• Uses fewer output tokens compared to Gemini 3.5 Flash;


• Requires fewer reasoning steps and tool calls;


• Supports multimodal scenarios;


• Is well-suited for building and deploying AI agents.

According to the Artificial Analysis Index, Gemini 3.6 Flash uses approximately 17% fewer output tokens than Gemini 3.5 Flash. In specific benchmarks, including DeepSWE, reductions can reach up to 65%.

What does this mean in practice? The model can deliver more accurate and useful results without unnecessary detail. It moves through action chains faster, repeats obvious steps less frequently, and works more efficiently with tools.

This is especially important for businesses: the fewer tokens and intermediate operations required to complete a task, the lower the total cost of the AI agent. Google also claims a reduced price for Gemini 3.6 Flash compared to the previous model — making it an attractive option for scaling automated solutions.

Best Use Cases for Gemini 3.6 Flash

Gemini 3.6 Flash can serve as a versatile AI assistant for complex workflows.

Programming

The model is suitable for:

• Code generation;


• Bug fixing and debugging;


• Refactoring;


• Explaining existing code;


• Writing tests;


• Working with large projects;


• Automating repetitive developer tasks.

Gemini 3.6 Flash is useful for both beginner programmers who need clear explanations and experienced developers handling complex, multi-stage tasks.

Text and Document Processing

The model helps analyze materials, structure information, highlight key points, and draft new text based on source data.

It enables you to:

• Create concise and detailed summaries;


• Compare documents;


• Draft business correspondence;


• Prepare reports and presentations;


• Edit and refine texts;


• Extract data from large volumes of information;


• Turn fragmented materials into a clear action plan.

Multimodal Tasks

Gemini 3.6 Flash goes beyond text processing. The model excels in scenarios requiring the analysis of multiple data types — such as text, images, and other visual assets.

This makes it convenient for analyzing screenshots, visual content, tables, diagrams, and documents, allowing you to solve tasks within a single AI interface without relying on separate services for each format.

Gemini 3.5 Flash-Lite: Maximum Speed and Efficiency

Gemini 3.5 Flash-Lite is the fastest and most cost-effective model in the Gemini 3.5 Flash class. It is built for scenarios where response speed, low cost, and high throughput are the top priorities.

According to Google and Artificial Analysis Index data, Gemini 3.5 Flash-Lite achieves generation speeds of up to 350 tokens per second.

The model is particularly well-suited for high-volume, rapid-turnaround tasks:

• Query classification;


• Customer support request processing;


• Short response generation;


• Translation and paraphrasing;


• Structured data extraction;


• List and catalog processing;


• Bulk text validation;


• Query routing to other AI models;


• Automation of simple business process stages.

Flash-Lite is an excellent choice for AI agents that need to execute quick individual steps: identify user intent, locate information, draft an initial response, or route the task to a more powerful model.

Which Model to Choose

The choice depends on your specific needs.

Choose Gemini 3.6 Flash if you need:

• Advanced reasoning capabilities;


• Programming assistance;


• Deep document analysis;


• Multimodal data handling;


• Building AI agents;


• Multi-step workflow execution;


• A balanced trade-off between quality, speed, and cost.

Choose Gemini 3.5 Flash-Lite if your priorities are:

• Maximum speed;


• Handling high request volumes;


• Short, concise responses;


• Cost-efficient AI usage;


• Bulk data processing;


• Simple automated operations.

You don't have to limit yourself to just one model. In real-world projects, they complement each other: Flash-Lite can quickly handle standard queries, while Gemini 3.6 Flash takes over tasks requiring deeper analysis.

Try New Gemini Models in vega.chat

vega.chat provides unified access to modern AI models in a single platform. It is ideal if you want to compare outputs across models, select the best fit for a specific task, and utilize top AI tools from Russia.

In vega.chat, you can:

• Test Gemini 3.6 Flash;


• Evaluate the speed of Gemini 3.5 Flash-Lite;


• Compare models side-by-side;


• Use AI for text, code, analysis, and document processing;


• Select the right model for personal or business needs;


• Avoid switching between multiple platforms.

The new Gemini models are available for testing now. Enter your prompt, upload a file, or describe a task to see how Gemini 3.6 Flash and Gemini 3.5 Flash-Lite perform in real-world applications.