Hy Image 3.5 Preview
Neural Network
Hy Image 3.5 Preview is a Tencent model for text-to-image generation and image editing via a single endpoint.
Specifications
- Input
- text, images
- Output
- images
- Resolution
- 1K, 2K, 4K
- Aspect ratio
- 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 16:9, 9:16, 21:9
import OpenAI from 'openai';
const openai = new OpenAI({
apiKey: '<your bothub access token>',
baseURL: 'https://openai.bothub.chat/v1'
});
async function main() {
const response = await openai.images.generate({
model: 'hy-image-v3.5-preview',
prompt: 'a white siamese cat',
n: 1,
size: '1024x1024'
});
const imageUrl = response.data[0].url;
}
main();
How it works Hy Image 3.5 Preview?
Hy Image 3.5 Preview is a unified Tencent model that creates images from text descriptions and edits existing ones. Text-to-image generation, image-to-image mode, and multi-step edits work through a single endpoint, accepting up to twenty references to maintain style and character consistency across frames. Via BotHub, the model is accessible from Russia without a VPN or foreign card: pay in rubles with a Russian card, tokens do not expire. Compare results easily with over 250 models in one window, and use a single OpenAI-compatible API to avoid rewriting integrations. Traffic is encrypted via AES-GCM; companies get contracts, invoices, EDI, and an admin panel with limits. Use cases: ad creatives and banners, product photo editing, concept art from mood boards, character consistency in series, and embedding image generation in your service.Frequently asked questions about Hy Image 3.5 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 generative image model from Tencent. It creates images from text descriptions and supports image-to-image mode, meaning it can use an uploaded reference. It is suitable for illustrations, concepts, covers, banners, social media visuals, and quick variations of an idea.
The output is a ready-made raster image that you can download from the BotHub chat or retrieve via API. Specific resolutions and aspect ratios depend on the request parameters, so it is easier to generate a couple of options and choose the one that fits your layout.
Step-by-step reasoning mode is not indicated in our data — the model is designed for image generation, not detailed analysis. If you need a chain of thought or calculations, use any text model in BotHub: switching takes one click, and there is no need to rewrite the integration.
Our data does not indicate support for function calling or structured output; the model is focused on visual results. For scenarios involving tools and strict response schemas, use BotHub text models via the unified OpenAI-compatible API — requests to them follow the same scheme.
Images — yes: the model supports image-to-image mode, so you can attach a reference and ask for a variation, style change, or composition refinement. Audio and video input are not indicated in our data — BotHub has separate specialized models for them.
We do not publish data on prompt languages, so it is more reliable to test it with your own task: describe the scene, style, and composition, and evaluate the result. If the wording is inaccurate, try clarifying the details or compare the output with another image model in BotHub.