Gemini 3.1 Flash Lite Batch
Neural Network
Gemini 3.1 Flash Lite is a Google multimodal model for fast, high-volume tasks; Gemini 3.1 Flash Lite API for agentic scenarios.
Specifications
- Input
- text, images, documents
- Output
- text
- Context size
- 1 048 576 tokens
- Max answer length
- 65 536 tokens
- Reasoning effort
- none, minimal, low, medium, high
- Capabilities
- reasoning, function calling
Prompt cost
(per 1M tokens)
Answer cost
(per 1M tokens)
Image prompt
(per 1K tokens)
How it works Gemini 3.1 Flash Lite Batch?
Gemini 3.1 Flash Lite is a Google Gemini family model designed for low latency and high-volume request streams. It accepts text, images, video, audio, and PDF, while the batch variant is intended for processing large volumes of data. The model was created for lightweight agentic scenarios where response speed and stream stability are more important than deep reasoning on a single task. In BotHub, it works from Russia without a VPN or foreign card, and you pay only for tokens used: no subscription is needed, and tokens do not expire. The unified BotHub API allows you to connect Gemini 3.1 Flash Lite to your service and switch to any of over 250 models at any time without rewriting the integration; chats are protected by AES-GCM encryption, and contracts, invoices, and electronic document management are available for companies. The model is most often used for mass classification and labeling of requests, extracting fields from scans and PDFs, summarizing documents and media files, drafting catalog descriptions, and quick steps within agentic chains.Frequently asked questions about Gemini 3.1 Flash Lite Batch
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.
This is a Google text model with a batch request processing mode. It accepts text, documents, and images as input, can reason before answering, and call functions. It is suitable for mass data processing: classification, extracting fields from documents, labeling, summarization, and consolidating large volumes into a structured result.
The maximum output is 65,536 tokens, which is enough for a long report, detailed documentation, or a batch of descriptions at once. The context window is 1,048,576 tokens, so the response can rely on voluminous source material: contracts, logs, exports, and entire sets of files.
Yes, the reasoning mode is confirmed in our data. The model breaks down the task into steps and only then provides a result — this helps with logic, calculations, and analyzing contradictory documents. In batch mode, such chains are convenient to run immediately on a large set of similar tasks.
Yes, function calling and structured JSON output are supported, so the model can be embedded into an agent or workflow and receive answers strictly according to your schema. BotHub uses a unified OpenAI-compatible API: you won't have to rewrite the integration when switching models.
Images and documents — yes: send a scan, PDF, or image and ask to extract data, describe content, or create a table. There are no notes in our data regarding audio and video, so do not count on them in advance — it is better to check with your own file.
We do not provide separate measurements for Russian, and we will not invent numbers. Gemini models generally work confidently with Russian, but it is more reliable to run a short test on your own data. Access to the model from Russia is without VPN or foreign cards, with payment via Russian cards.