GigaChat-Max-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-Max-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-Max-preview',
stream: true
});
for await (const chunk of stream) {
const part: string | null = chunk.choices[0].delta?.content ?? null;
}
}
main();
How it works GigaChat-Max-preview?
Frequently asked questions about GigaChat-Max-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-Max-preview is a text model: text in, text out. It is suitable for writing and editing materials, summarizing long documents, analytics, working with code, and answering questions. The 128,000-token context allows you to upload a voluminous document or an entire correspondence and discuss them.
Up to 128,000 tokens in a single response — the output limit here matches the context window size. This is enough for a large report, detailed documentation, or an extensive analysis. Keep in mind that the sent text and the response share the same window, so leave some buffer.
There is no specific note about a reasoning mode in our data, so we do not promise it in advance. However, you can request a breakdown directly in the prompt: 'break down the task step by step', 'reasoning first, then conclusion'. Test this scenario with your task.
There is no note about function calling or strict JSON mode in our data — this is not a denial, just no confirmation. If you need a guaranteed structured response, describe the schema directly in the request and validate the result, or switch to a model with confirmed support in BotHub.
According to our data, the model is marked as text-only: input and output are text; there are no notes about accepting images, audio, or video. For such tasks, switch to a multimodal model in the same BotHub window: the API is unified, so you won't have to rewrite the integration.
This is a Russian-developed model, and Russian is its native language: cases, stylistics, business, and conversational phrasing are handled confidently. There are no separate quality measurements in our data, so it is easier to check for yourself — send your typical text and compare the response with another model.