deep-research-preview-04-2026
Neural Network
Specifications
- Input
- text, images, documents
- Output
- text
- Context size
- 1 048 576 tokens
- Max answer length
- 32 000 tokens
- Capabilities
- 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: 'deep-research-preview-04-2026',
});
}
// Async - Text generation
async function main() {
const stream = await openai.chat.completions.create({
messages: [{ role: 'user', content: 'Say this is a test' }],
model: 'deep-research-preview-04-2026',
stream: true
});
for await (const chunk of stream) {
const part: string | null = chunk.choices[0].delta?.content ?? null;
}
}
main();
How it works deep-research-preview-04-2026?
Frequently asked questions about deep-research-preview-04-2026
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 Google with a context window of 65,536 tokens. It is suitable for in-depth analysis of large materials: analytical reviews, reports, working with documents and images, preparing structured summaries, as well as for code, refactoring, and tasks requiring sequential reasoning.
The output limit is up to 32,000 tokens per response. This is enough for a voluminous analytical breakdown, detailed documentation, a long report, or a large code fragment without manually splitting the task into parts. If necessary, continue the generation with the next message.
In our data, a separate reasoning mode for this model is not noted, so we will not guarantee it. In practice, a simple trick helps: ask in the prompt to break down the task step-by-step, verify facts, and only then provide the final conclusion.
In our data, function calling and structured output are not noted for this model. If the response format is critical, describe the required schema directly in the prompt. BotHub provides a unified OpenAI-compatible API, so you can switch to a model with confirmed function calling without rewriting the integration.
Yes, images and documents are accepted as input in addition to text: you can ask to analyze a screenshot, diagram, table, or PDF and receive a text analysis. Audio and video are not listed as supported inputs for this model; for those, choose specialized models from the catalog.
Google models traditionally work with Russian confidently, but we do not have official quality data for this specific version. Test it with your task: in BotHub, access is open from Russia without a VPN or foreign card, and payment is in rubles based on actual token usage.