GigaChat-Pro-preview
Neural Network
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-Pro-preview',
});
}
// Async - Text generation
async function main() {
const stream = await openai.chat.completions.create({
messages: [{ role: 'user', content: 'Say this is a test' }],
model: 'GigaChat-Pro-preview',
stream: true
});
for await (const chunk of stream) {
const part: string | null = chunk.choices[0].delta?.content ?? null;
}
}
main();
How it works GigaChat-Pro-preview?
Frequently asked questions about GigaChat-Pro-preview
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-Pro-preview works with text: it writes and edits materials, summarizes long documents, answers questions, and helps with code—analysis, refactoring, debugging. The 128,000-token context allows you to work with large files, correspondence, or technical documentation in their entirety, without splitting them into parts.
The output limit is 128,000 tokens per response, the same as the context window. This is enough for a long report, a series of articles, or a voluminous code module. In practice, it is more convenient to request text in parts: this makes it easier to edit and control the structure.
In our data, a separate reasoning mode for this model is not noted, so we cannot confirm it. However, you can ask it to break down a task step-by-step directly in the prompt: this usually significantly improves answers regarding logic, calculations, and STEM.
Function calling and strict JSON output are not marked in our data — which means we neither confirm nor deny them. The fastest way to check is to send a request via the unified OpenAI-compatible BotHub API and see how the model responds to your schema.
According to our data, the model is declared as text-only: TEXT_TO_TEXT. The acceptance of images, audio, and video is not separately noted, so focus on text-based scenarios. If you need to work with images or sound, multimodal models are available nearby in BotHub — switching happens in the same window.
GigaChat is a Sber development, and the line was originally created for a Russian-speaking audience, so the model works confidently with tasks in Russian. There are no specific quality ratings in our data, so test it with your own texts: access in BotHub is open without a VPN or foreign card.