YandexGPT Lite
Neural Network
Specifications
- Input
- text, documents
- Output
- text
- Context size
- 32 000 tokens
- Max answer length
- 32 768 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: 'yandexgpt-lite',
});
}
// Async - Text generation
async function main() {
const stream = await openai.chat.completions.create({
messages: [{ role: 'user', content: 'Say this is a test' }],
model: 'yandexgpt-lite',
stream: true
});
for await (const chunk of stream) {
const part: string | null = chunk.choices[0].delta?.content ?? null;
}
}
main();
How it works YandexGPT Lite?
Frequently asked questions about YandexGPT Lite
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 from Yandex for everyday text work: summaries, rewriting and editing, answering questions, drafting emails and descriptions. It accepts documents as input, so it is suitable for analyzing files and extracting the main points from long materials.
Up to 32,768 tokens in a single response — this is enough for a detailed article, a large report, or a series of descriptions in one request. The context window is 32,000 tokens, which accommodates your document, instructions, and dialogue history.
In our data, a separate reasoning mode for this model is not noted. However, you can get a step-by-step breakdown with a standard prompt: ask it to write out the logic step-by-step. If you need detailed chains of thought, reasoning models are available nearby in BotHub — switching takes just one click.
In our data, support for function calling and strict structured output is not noted, so we cannot guarantee it. In practice, the model usually responds according to a given template if you specify the format in the prompt. For reliable schemes, choose a model with confirmed structured output.
The model works with text and documents: you upload a file, and it summarizes it, answers questions about the content, or rewrites fragments. For images, audio, and video, there are separate multimodal models in BotHub — they are available in the same window and via a single API.
The model was created by Yandex and is initially focused on Russian-language tasks, which is noticeable in editing, summarizing, and business correspondence. The best way to check is to compare its response with other models in BotHub using your text: access from Russia without a VPN or foreign card.