Site map
- Home
- Capability timeline
- Archives
- Compare models by task
- Midjourney Updates
- Grok Imagine Video 1.5
- Nano Banana
- Seedance
- Seedream
- Wan
- GPT Image
- Gemini Omni
- Veo
- Kling
- MiniMax Hailuo
- Ideogram
- FLUX
- Recraft
- Grok Imagine Image 2.0
- Runway Gen-4 References
- Runway Gen-4.5
- Runway Aleph
Change articles
Adobe Photoshop
Autodesk Forma
- What Forma's Indoor Daylight public beta opened up ·
- From a Forma floor plan to an AutoCAD drawing ·
- Planning vertical circulation with native stairs in Forma ·
Canva
- Affinity brings its unified design app to iPad ·
- What stayed editable after Canva Brush revised a design ·
- Where Canva Craft fits in design work ·
- Canva connector for Meta Muse ·
- An Affinity file reaches a Cavalry composition ·
- What Canva AI 2.0 made editable after the first prompt ·
ChatGPT image
- ChatGPT Images 2.6 in practice ·
- ChatGPT Images 2.6 brought animated GIFs into the chat ·
- What ChatGPT Images 2.5 changed when editing an image ·
- How ChatGPT Images 2.0 changed the infographic brief ·
- What held up across edits in December's ChatGPT Images ·
- ChatGPT added an image-generation switch for parents of linked teens ·
Claude web design
- Claude Code adds a deadline notice for standalone Claude Design ·
- Claude Code tells the Artifact tool to update shared artifacts in place ·
Gemini image
- Gemini Live turns a phone camera view into an image edit ·
- What Nano Banana could keep intact during a Gemini edit ·
- What early Imagen 4 users learned about lettering and layout ·
Grok
- Grok Imagine Image 2.0 becomes the consumer Quality Mode ·
- Making a usable shot with Grok Imagine Video 1.5 ·
- Grok Imagine’s legacy Quality Mode API ·
- xAI raised Grok Imagine's image-edit API limit from three source images to five
- Grok Imagine added 21:9 and 5:2 to its image API
- What the January Grok Imagine API launch made available
- Grok Imagine Video 1.5 added prompt-only and reference-guided video in July
Higgsfield
- Drawing a series with Higgsfield Mangaka ·
- Higgsfield Genjutsu Object Swap ·
- Cinema Studio 3.5 gave a sequence one visual setup ·
- What Higgsfield Soul 2.0 carries from a fashion reference ·
- Higgsfield Angles v2 ·
- Higgsfield introduces SOUL for fashion and editorial stills
- Higgsfield Popcorn turned a reference still into a sequence of shots
Lovable
- How Lovable's Figma plugin handed a design to the app builder ·
- Lovable expands Visual edits to multiple elements and mixed-content text ·
Midjourney
- What the Midjourney V8.2 edit model carried from one image to the next ·
- How Midjourney V8.2 combined separate image references ·
- V8 Alpha image-generation speed ·
- V8 Alpha native HD and quality 4 ·
- Midjourney Video V1 made both ends of a shot selectable ·
- Midjourney Video V1 brought motion into the image workflow ·
- What changed when Midjourney made V7 the default ·
- What Midjourney V7 Omni Reference carried between scenes ·
Runway
- Runway Aleph 2.0 lets editors preview a change before re-rendering the shot ·
- Seedance 2.0 arrives in Runway ·
- Comparing video models inside Runway ·
- Runway Gen-4.5 gains a first-frame image input ·
- Runway opens Gen-4.5 text-to-video to all paid plans ·
- Runway opens Aleph video editing to all paid plans ·
- Runway updates Gen-4 References for object consistency ·
Midjourney capability changes
- Selected-pixel editing
- V8.2 Edit Model enters community testing
- Draft candidates become full-size jobs
- V8.1 becomes default
- SD-to-HD rerun on the web
- SD temporarily returns as default
- V8.1 Alpha · rerun SD as HD
- Native 2K rendering opens in Alpha
- 720p access expands
- Native 720p video
- Looping and end-frame control
- Video V1 · animate a still
- V7 Omni Reference in the default model
- V7 Omni Reference opens for testing
- Select and remove an object
- Region editing opens to uploaded images
- Unified web editor
- Vary Region appears as Repaint
- V6 Character Reference test