Tencent Releases Preview Of Hy Image3.5 Generation Model
Tencent is pricing and distributing a preview model across its own cloud plus partner tools at once, so the terms of access matter as much as the image quality claims.
Reporting from 1 source: GIGAZINE.
Tencent released a preview version of Hy Image3.5, an image model that handles text-to-image and image-to-image generation in one system with a maximum output size of 2K. Tencent says human evaluations rated its professional-grade accuracy 30 percent better than Hy Image3.0. It can also edit images. Access runs through the Tencent Cloud API at $0.024 per generated image, with referencing images free.
The preview is not confined to one surface. Access runs through the Tencent Cloud API at $0.024 per generated image, and the system charges only for images it produces, with referencing images free.
For a limited two-week period it is also free on the AI creative tool Miora and the AI video production tool OnSolo, and it is available through GMI Cloud, an AI-native inference cloud supported by NVIDIA. ComfyUI users can reach it too, though that route needs account registration and adds a charge for each process.
Synthesized by Yomimono from the 1 cited source below, including Japanese-language reporting where cited, then editorially reviewed before publishing.