Gemini 3.5 Flash Lite
Neural Network
Gemini 3.5 Flash Lite is a high-efficiency Google model for sub-agents performing specific tasks in multi-agent systems.
Specifications
- Input
- text, images, documents
- Output
- text
- Context size
- 1 048 576 tokens
- Max answer length
- 65 536 tokens
- Reasoning effort
- minimal, low, medium, high
- Capabilities
- reasoning, function calling, Batch API
Prompt cost
(per 1M tokens)
Answer cost
(per 1M tokens)
Image prompt
(per 1K tokens)
How it works Gemini 3.5 Flash Lite?
Gemini 3.5 Flash Lite is a Google model designed for high efficiency and advanced agent capabilities. It is tailored for sub-agents performing specific tasks within complex multi-agent processes that require fast, predictable execution of repetitive steps. On BotHub, the model is accessible from Russia without a VPN or foreign card: you pay in rubles for tokens used, with no subscription, and unused tokens do not expire. A unified OpenAI-compatible API allows you to connect Gemini 3.5 Flash Lite to your product and switch models without rewriting the integration. Traffic is encrypted via AES-GCM, chats are not saved, and companies can access contracts, invoices, electronic document management (EDO), and an admin panel with limits. It is useful for building pipelines where a strong model plans and Gemini 3.5 Flash Lite executes individual steps; labeling and classifying large datasets; extracting structured fields from documents; categorizing incoming requests and drafting responses; and automating routine operations within agent scenarios.Frequently asked questions about Gemini 3.5 Flash Lite
You can use generated results for commercial purposes. You own all rights to the content you create. The only restriction: make sure your prompt does not include copyrighted third-party material. You are responsible for respecting the rights to any input data.
gemini-3.5-flash-lite works with text, accepts images and documents as input, including native processing. It is suitable for processing large volumes of data, parsing files and reports, extracting meaning from scans and screenshots, supporting dialogue scenarios, and batch tasks.
Up to 65,536 tokens per response—that's dozens of pages. Enough to get a large document, detailed analysis, or long summary in one go without splitting. The context window is 1,048,576 tokens, so input data can also be voluminous.
Yes, reasoning before answering is confirmed by our data: the model can break down a task into steps before formulating a conclusion. This helps with calculations, logic puzzles, parsing contradictory conditions, and multi-step instructions where a correct result is more important than a fast response.
Yes, the model supports function calling and structured JSON output. You describe the schema and tools, and the model returns a predictable object that can be passed directly into code. On BotHub, this works via a unified OpenAI-compatible API, so switching to another model does not break the integration.
According to our data, images and documents are accepted as input, including native file processing: you can send a screenshot, diagram, scan, or PDF and ask for an analysis. We have no notes on audio and video, so we do not promise them in advance—please check the required format with a small request.
We have no data on specific language support, so we will not make claims about the quality of Russian—please evaluate this with your typical request. What is known for sure: access to the model on BotHub works from Russia without a VPN or foreign card, with payment via Russian cards in rubles.