Qwen-Image 2.1: generation, editing and transparency in one model
Alibaba detailed the Pro and Turbo APIs for its image model: from $0.014 an image, up to 10 references and a real alpha channel.
Images
5 entries
Canva Craft: Canva’s first in-house image model, trained on templates
Canva released a model trained on human-made designs rather than photos. Results can become editable layers.
FLUX 3 Image Klein: an open image model that runs on 8 GB of VRAM
Black Forest Labs released a 5.6B-parameter model with 4K, editing and ten reference images. Commercial use is free for small businesses.
Midjourney tests Thinking Mode for better prompt accuracy and type
On the Alpha site, you can rerun an image in an experimental “thinking” mode. The company wants feedback before deciding on a wider release.
Nano Banana 2.1: better prompt accuracy, text and consistent characters
Google’s new image generation and editing model is generally available in the Gemini API. The previous version is already deprecated.