Hy Image3.5 preview is live. 🚀 Professional-grade image generation, +30% win…
- Source
- x.com
- Date
Hy Image3.5 preview is live. 🚀 Professional-grade image generation, +30% win rate in human eval vs Hy Image3.0 Both Text to image & Image to image available. Up to 2K. Better Consistency. API: https://t.co/5BvR2SdI6Q Priced for everyone. $0.024 per image on Tencent Cloud API. We only charge for what we generate — your reference images are free. Two weeks free (Only in OnSolo and Miora): Miora https://t.co/UxwBScI64O OnSolo https://t.co/iXnNNg6U9k Try it and tell us where it breaks.
Context
Tencent's Hunyuan team put a preview of Hy Image3.5 live, its next image generation and editing model after Hy Image 3.0. Tencent reports a 30% win rate in human evaluations against the 3.0 version, output up to 2K resolution, and both text-to-image and image-to-image modes.
Pricing on Tencent's Cloud API is $0.024 per generated image, and Tencent says it only charges for images actually produced, not for reference images supplied as input. The 30% win rate is Tencent's own claim, not an independently verified , so how the model would score on a third-party leaderboard is not established by this announcement.
- eval — A repeatable test for AI quality — a set of tasks plus scoring — used the way software teams use test suites, because model output is too variable to judge by eyeballing.
- benchmark — A standard public test set for comparing AI models — the shared scoreboards behind every "model X beats model Y" claim.
Checking sign-in…
Loading comments…