Gemini 3.1 Pro Preview Custom Tools
Neural Network
Gemini 3.1 Pro Preview Custom Tools is a Google model with precise function calling for agent scenarios and coding tasks.
Max answer length
(in tokens)
Context size
(in tokens)
Prompt cost
(per 1M tokens)
Answer cost
(per 1M tokens)
Image prompt
(per 1K tokens)
How it works Gemini 3.1 Pro Preview Custom Tools?
Gemini 3.1 Pro Preview Custom Tools is a variant of Google's Gemini 3.1 Pro with refined tool selection: the model is less likely to default to generic bash when a specialized third-party tool is more effective. This distinguishes it from the base version, resulting in shorter and more predictable call chains. BotHub provides access without VPN or foreign cards, with payment in rubles via Russian cards on a pay-as-you-go basis; tokens do not expire. Access over 250 models in one window, and a unified OpenAI-compatible API allows switching between them without rewriting integrations. Chats are encrypted via AES-GCM. For companies, we offer contracts, invoices, electronic document management (EDO), and an admin panel with limits. Useful if you are building an agent with external tools, automating coding tasks (refactoring, debugging, reviews), creating function-calling scenarios where call discipline is critical, or prototyping a multi-model service using a single key.Frequently asked questions about Gemini 3.1 Pro Preview Custom Tools
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.
The model accepts text, documents, and images, and responds with text. The context window is 1,048,576 tokens, so you can provide long reports, large code files, and collections of materials. It is suitable for analytics, coding, document analysis, and tasks requiring reasoning and function calling.
The output limit is 65,536 tokens per response. This is enough for voluminous material: detailed analysis, technical documentation, a large code module, or a long structured report. If the task is larger, split it into parts and continue generation with the next request in the same dialogue.
Yes, the model's reasoning before answering is confirmed. It builds a sequence of steps before formulating a result, which significantly helps with mathematics, logic, code debugging, and multi-step document analysis. For complex queries, the response may take longer to prepare, but the conclusions are more well-founded.
Yes, both function calling and structured JSON output according to your schema are supported. This is the foundation for agents, integrations with internal services, and databases. In BotHub, the model is available via a unified OpenAI-compatible API, so you can replace it with another without rewriting your integration.
You can send images and documents: the model analyzes screenshots, diagrams, tables, and files along with a text query. Audio and video input is not noted in our data, so it is better to test this scenario with a small request before integration. The model always returns the result as text.
There are no separate language measurements in our data, so we cannot make exact promises—evaluate the results with your own materials. The Gemini Pro line generally works confidently with multilingual texts. In BotHub, the model is accessible from Russia without a VPN, with payment via Russian cards on a pay-as-you-go basis.