A number of image generation models are being released. This website doesn’t go deep into details but aims to at least list them.
Anima is a specialized, open-weights text-to-image AI model with 2 billion parameters, created by CircleStone Labs in collaboration with Comfy Org. Unlike general-purpose models, it is specifically fine-tuned for high-quality anime, manga, and non-photorealistic artistic illustrations.
License allows commercial use of generated images.
Ideogram is a powerful model but it is released under a non-commercial license
HF:microsoft/Lens 3.8B-parameter foundational text-to-image model designed for efficient training and fast high-resolution generation. … FLUX.2 semantic VAE
20B parameters, Apache 2.0
And it has the natural look which I personally prefer than the over “RLd” [extremely artificial] look that Z-Image has.
Haven’t noticed that, z-image looks very realistic to me
I call that distill look; but I agree zit looks less AI than other models
While previously LongCat referred only to a video generation model now we also have image generation and editing models under the same name
Ovis 7B image generation model released: GH, HF; the model is likened to and compared with z-image-turbo