Specifications
- Input
- text, images, documents
- Output
- text
- Context size
- 1 048 576 tokens
- Max answer length
- 32 000 tokens
- Capabilities
- Batch API
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: 'genie-4',
});
}
// Async - Text generation
async function main() {
const stream = await openai.chat.completions.create({
messages: [{ role: 'user', content: 'Say this is a test' }],
model: 'genie-4',
stream: true
});
for await (const chunk of stream) {
const part: string | null = chunk.choices[0].delta?.content ?? null;
}
}
main();
How it works genie-4?
Frequently asked questions about genie-4
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.
genie-4 works with text: it accepts text prompts, images, and documents, and responds with text. A context window of up to 1,048,576 tokens allows you to upload large materials in their entirety, and batch mode is useful when there are many requests at once. Through BotHub, the model is available from Russia — without a VPN or a foreign card.
In a single response, genie-4 generates up to 32,000 tokens, and the input context reaches 1,048,576 tokens. This is enough for long analyses and working with voluminous documents: you can upload the material in its entirety without cutting it into pieces. The exact volume in words depends on the text itself.
There is no separate confirmation in our data that genie-4 spends a step on reasoning before answering — the corresponding mark is missing. Therefore, we will not promise this in advance: it is easier to check the behavior on long logical tasks using your own prompts in the BotHub interface, paying only for the tokens consumed.
Function calling and structured JSON output are not marked in our data, so we cannot guarantee them. If a strict response format is important for your scenario, check it with a test request: in BotHub, you pay for tokens as you go, no subscription is needed, and tokens do not expire.
genie-4 accepts images and documents as input — this is confirmed by our data. We have no notes regarding audio and video, so we will not promise their support. If your task is based on images or files, the model will work; for other formats, it is worth checking in practice.
There is no information in our data about the quality of Russian specifically for genie-4, so we will not promise fluent Russian. It is more practical to give the model a few of your own tasks in Russian and see the result: payment in BotHub is based on actual token usage, without a subscription and without a foreign card.