Specifications
- Input
- text
- Output
- text
- Context size
- 128 000 tokens
- Max answer length
- 128 000 tokens
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: 'GigaChat-2',
});
}
// Async - Text generation
async function main() {
const stream = await openai.chat.completions.create({
messages: [{ role: 'user', content: 'Say this is a test' }],
model: 'GigaChat-2',
stream: true
});
for await (const chunk of stream) {
const part: string | null = chunk.choices[0].delta?.content ?? null;
}
}
main();
How it works GigaChat-2?
Frequently asked questions about GigaChat-2
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.
GigaChat-2 from Sber works with text: it writes and edits articles, emails, and descriptions, summarizes long documents, answers questions, and helps with code and refactoring. A context window of up to 128,000 tokens allows you to upload entire reports and discuss them in full.
The output limit is up to 128,000 tokens, the same as the context window. In practice, this is enough for a voluminous longread, a detailed instruction, a large block of code, or a comprehensive document analysis without needing to break the task into parts.
A separate reasoning mode is not noted in our data. However, you can ask the model to break down the task step-by-step directly in the prompt — this technique usually improves answers to logic and calculation tasks. If you specifically need reasoning, switch to a reasoning model in BotHub.
These capabilities are not noted in our data, so do not count on them in advance. You can specify a strict response format directly in the prompt with a structure example. If tools are critical, models with confirmed function calling are available in BotHub — you won't have to rewrite the integration.
In our data, the model is marked as text-only: input and output are text. For images, audio, and video, use specialized BotHub models and switch between them in one window without changing your account or payment method.
This is a Russian development by Sber, and Russian is one of its primary working languages. The model confidently maintains style, grammatical cases, and business phrasing. It is easier to check the quality for your tasks in practice: compare the answers with other models in BotHub without a VPN or foreign card.