Image generation
Image models, control, editing, inpainting, identities, consistency and upscaling
- discussionWhat is the most reliable way to keep a character and set visually consistent across 50+ storyboard panels with open models?
- sourceRunway Academy: Gen-4 References
- sourceHow to Keep Your AI Character Consistent Across Every Generation: Soul ID Explained
- sourceMastering AI Character Consistency in Video and Images (Soul ID)
- sourceNetflix's Moria hybrid campaign by Vertex CGI with Higgsfield
- sourceOne of One: Maradona Official brand world from concept to sold out
- sourcexinntao/Real-ESRGAN
- sourceToTheBeginning/PuLID
- sourceBrilliance cuts production costs 80% with Marketing Studio
- sourceTencentARC/PhotoMaker
- sourcetencent-ailab/IP-Adapter
- sourceFLUX.1 Kontext: Flow Matching for In-Context Image Generation and Editing in Latent Space
- sourceQwenLM/Qwen-Image
- sourceInstantX/InstantID
- sourceStable Diffusion XL Base 1.0
- sourceHugging Face Diffusion Models Course (diffusion-models-class)
- sourceHiDream-ai/HiDream-I1
- sourceStable Diffusion 3.5 Large
- sourceQwen-Image-Edit-2511
- sourceDiffusers guide: inpainting (with image-to-image chaining)
- sourceFLUX.1 Kontext [dev] (instruction image editing)
- sourceFanghua-Yu/SUPIR
- sourcedjghosh13/geneval (GenEval)
- sourceblack-forest-labs/flux2 (FLUX.2 inference code)
- sourceblack-forest-labs/flux (FLUX.1 inference code)
- sourceBlack Forest Labs prompting reference (FLUX.1 Kontext, FLUX.2, FLUX 3 Image)
- sourceVertexCGI builds a viral NBA Finals campaign in a five-day sprint
- sourceAdQUA and Boosterz: Korean agencies report 97% AI production cost cut
- sourceStoryGen: Intelligent Grimm, open-ended visual storytelling via latent diffusion (CVPR 2024)
- sourceStoryDiffusion: consistent self-attention for long-range image and video generation
- sourceConsiStory: training-free consistent text-to-image generation (NVIDIA)