GigaChat-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-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-preview',
stream: true
});
for await (const chunk of stream) {
const part: string | null = chunk.choices[0].delta?.content ?? null;
}
}
main();
How it works GigaChat-preview?
Frequently asked questions about GigaChat-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-preview works with text: it writes and edits materials, summarizes long documents, answers questions based on provided text, and helps with code and business correspondence. The 128,000-token context allows you to keep an entire contract, article, or technical specification in the dialogue.
The output limit is 128,000 tokens, the same as the context window. This is enough for very voluminous text: a detailed report, a chapter, or a series of related materials. Keep in mind that the request and response share the same window, so long input data reduces the space available for the response.
A separate reasoning mode is not noted in our data, so we will not promise it. However, you can always ask the model to write out the solution step-by-step, verify intermediate conclusions, and explain the logic — this usually significantly improves the quality of the answer for complex tasks.
Function calling and strict JSON output are not noted in our model card, so you shouldn't rely on them in production without testing. You can request a JSON response directly in the prompt, but it is better to test the result on your scenarios via the unified BotHub API.
The model is declared as text-only: input of images, audio, and video is not noted in our data. If the task requires analyzing an image or transcribing a recording, switch to a suitable model in the same BotHub window — you won't have to rewrite the integration and API.
GigaChat is a Russian development, and Russian is its primary language: the model works with it natively, without intermediate translation. How accurately it matches your style is easier to check with real texts — in BotHub, this is available from Russia, without a VPN or foreign card, with payment in rubles.