gemini-3.5-transcribe-preview
Neural Network
Specifications
- Input
- text, images, documents
- Output
- text
- Context size
- 65 536 tokens
- Max answer length
- 32 000 tokens
- Reasoning effort
- low, medium, high
- Capabilities
- reasoning, function calling
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-preview',
});
}
// 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-preview',
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-preview?
Frequently asked questions about gemini-3.5-transcribe-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.
This is a model from google-gemini that accepts text, documents, and images, and responds with text. It is suitable for transcribing and structuring the content of files and scans, data extraction, analytics, and working with code. The context is 65,536 tokens, so voluminous materials fit into a single request.
In a single response, the model outputs up to 32,000 tokens — this is enough for a long document transcription, a detailed analysis, or a large code snippet. If the final material is larger, break the task into parts and continue generation with subsequent requests, relying on the overall context of the dialogue.
Yes, the model has a reasoning mode: before answering, it sequentially analyzes the condition and only then formulates the result. This significantly helps with tasks involving complex logic, document structure analysis, mathematics, and code debugging, where a verified answer is more important than a fast one.
Yes, the model supports function calling and structured JSON output. You describe the schema and tools, and it returns an object ready for parsing — convenient for data extraction pipelines, integrations, and agents. In BotHub, this is available via a unified OpenAI-compatible API.
According to our data, the model accepts images and documents in addition to text: you can send a scan, a photo of a page, a diagram, or a PDF and get a text result. Input support for audio and video is not noted, so focus on documents and images.
Models in the Gemini family from google-gemini traditionally work with Russian at a good level — both in understanding the request and in parsing text from images and documents. We do not have exact data on prompt and output languages, so please check the quality for your specific task in BotHub.