gemini-3.5-transcribe
Neural Network
Specifications
- Input
- text, images, documents
- Output
- text
- Context size
- 98 304 tokens
- Max answer length
- 32 000 tokens
- Reasoning effort
- low, medium, high
- Capabilities
- reasoning, function calling, Batch API
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: 'gemini-3.5-transcribe',
});
}
// Async - Text generation
async function main() {
const stream = await openai.chat.completions.create({
messages: [{ role: 'user', content: 'Say this is a test' }],
model: 'gemini-3.5-transcribe',
stream: true
});
for await (const chunk of stream) {
const part: string | null = chunk.choices[0].delta?.content ?? null;
}
}
main();
How it works gemini-3.5-transcribe?
Frequently asked questions about gemini-3.5-transcribe
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.
Google's gemini-3.5-transcribe works with text, documents, and images, turning them into structured text. The model reasons before answering, calls functions, and maintains a context of up to 65,536 tokens, making it suitable for transcribing file content, data extraction, and analytics. Access to BotHub from Russia without a VPN.
In a single response, the model outputs up to 32,000 tokens — roughly a long article or a detailed transcript of dozens of pages. If the material is larger, break the work into parts and submit them sequentially: the 65,536-token total context allows for this.
Yes, reasoning before answering is confirmed: the model breaks down the task step-by-step before providing a result. This is noticeable with complex documents, multi-step instructions, and tasks where it is important to connect disparate text fragments and reach a reasoned conclusion.
Yes, both. The model calls functions and returns structured JSON, which is convenient for storing in a database or passing along a pipeline. BotHub uses a unified OpenAI-compatible API: connect once and switch models without rewriting the integration.
Images and documents — yes, they are confirmed in our data: you can send a scan, PDF, or photo and get a text analysis. We have no notes regarding audio and video, so please check the available formats in the interface before setting up your process.
Our data does not specify language characteristics, so we cannot make any promises. The easiest way to check is in practice: send your typical document or task to BotHub and compare the result with other models — switching takes a couple of seconds, payment with Russian cards is supported.