Gemini 3.6 Flash Batch
Neural Network
Gemini 3.6 Flash Batch is a Google model for code, agent scenarios, and web development, available via the BotHub API.
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
Prompt cost
(per 1M tokens)
Answer cost
(per 1M tokens)
Image prompt
(per 1K tokens)
How it works Gemini 3.6 Flash Batch?
Gemini 3.6 Flash Batch is a highly efficient Google model for programming, agent scenarios, and web service/application development. It focuses on result accuracy: the output is more polished and requires fewer edits. The batch mode is designed for tasks requiring bulk responses rather than real-time interaction. Via BotHub, the model works without VPN or foreign cards; payment is in rubles via Russian cards, charged per used token (tokens do not expire). Access over 250 models in one window for easy comparison and switching. A unified OpenAI-compatible API allows changing models in a multi-model system without rewriting integrations. Data is encrypted via AES-GCM. For companies, BotHub provides contracts, invoices, electronic document management, and an admin panel with limits. It is suitable for processing task batches from trackers, rewriting legacy modules, prototyping web interfaces from text descriptions, serving as an agent core for repository management, or reviewing large commit sets.Frequently asked questions about Gemini 3.6 Flash 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 1,048,576 token context window. It analyzes long documents and images, writes and edits code, solves reasoning tasks, and calls external functions. Suitable for analyzing large files, user support, batch data processing, and multi-model services via a unified API.
The model outputs up to 65,536 tokens per response—enough for a large document, detailed report, or a major code module. If the task is larger, split it into parts: the 1,048,576 token context allows keeping the entire history in the dialogue.
Yes, reasoning before answering is confirmed in our data. The model breaks down tasks into steps before providing a result, which is noticeable in math, logic, code debugging, and analyzing contradictory documents. The benefit of this mode is smaller for simple queries than for complex ones.
Yes. Function calling and structured JSON output are confirmed, so the model can be connected to your tools, databases, and external services to receive predictable responses based on a schema. In BotHub, this works via a unified OpenAI-compatible API, so the model can be changed without rewriting integrations.
Images and documents—yes, they are indicated in our data: send a scan, PDF, table, or screenshot and get a text analysis. There is no note regarding audio and video, so we won't promise support in advance—it's easier to test with your own file.
Our data contains no information about prompt and response languages, so we will not make claims about the quality of Russian. The most reliable way is to check for yourself: send a typical query and compare the response with another model in the same BotHub window; switching takes seconds.