Gemini 4 Carbon
Neural Network
Google's Gemini 4 Carbon is a model for code, agentic scenarios, and complex reasoning, accepting text, images, audio, video, and files.
Specifications
- Input
- text, images, documents
- Output
- text
- Context size
- 1 048 576 tokens
- Max answer length
- 943 718 tokens
- Reasoning effort
- low, medium, high
- Capabilities
- reasoning, function calling
Prompt cost
(per 1M tokens)
Answer cost
(per 1M tokens)
Image prompt
(per 1K tokens)
import OpenAI from 'openai';
const openai = new OpenAI({
apiKey: '<your bothub access token>',
baseURL: 'https://openai.bothub.chat/v1'
});
// Sync - Text generation
async function main() {
const chatCompletion = await openai.chat.completions.create({
messages: [{ role: 'user', content: 'Say this is a test' }],
model: 'gemini-4-carbon:batch',
});
}
// Async - Text generation
async function main() {
const stream = await openai.chat.completions.create({
messages: [{ role: 'user', content: 'Say this is a test' }],
model: 'gemini-4-carbon:batch',
stream: true
});
for await (const chunk of stream) {
const part: string | null = chunk.choices[0].delta?.content ?? null;
}
}
main();
How it works Gemini 4 Carbon?
Gemini 4 Carbon is Google's advanced Gemini 4 model for programming, agentic scenarios, and complex reasoning. It accepts text, images, audio, video, and files as input. The provider highlights support for long responses, making it ideal for tasks requiring detailed output in a single pass. Its strengths include practical coding and step-by-step actions: analyzing project structure, debugging, step-by-step refactoring, planning changes, and verifying results. Via BotHub, this model is accessible from Russia without a VPN or foreign card: pay in rubles as you go, and tokens do not expire. Other models are available in the same window for easy comparison without rewriting integrations. For your services, there is a unified OpenAI-compatible API with data encryption and corporate terms including contracts, invoices, and limits. Scenarios: code development and debugging, agentic pipelines, engineering and scientific tasks, analyzing long documents and multimedia, and preparing technical documentation.Frequently asked questions about Gemini 4 Carbon
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 text model with a very large context: it accepts text, images, and documents, can reason before answering, and call functions. It is suitable for analyzing long documents, working with code, analytics, and tasks requiring structured JSON output via the BotHub API.
In a single response, the model can output up to 943,718 tokens, and the input context reaches 1,048,576 tokens. This allows for obtaining voluminous materials without splitting them into parts. The final volume depends on your request and how the model is configured.
Yes, the reasoning mode before answering is confirmed in our data. It can first process the task logic and then formulate the result. This is useful for multi-step calculations, analyzing conditions, and code verification, where the process of solving is as important as the result.
Yes, both capabilities are marked as confirmed in our data. The model supports function calling, so it can be connected to external tools and data sources, and it can return structured output in JSON. This is convenient for integrations via the BotHub API and process automation.
The model accepts images and documents — this is indicated in our data. There are no markings for audio and video, so we cannot confirm their acceptance. If you need audio or video processing, there are other models in the BotHub catalog, and you can switch to them in the same window.
There is no separate assessment of Russian language quality in our data, so we will not make specific promises. The model works with text and documents, but the result should be checked against your own tasks. You can compare options in the same BotHub interface by switching between models.