Training Text-to-Image Models Just Got 3.6 Times Faster
The efficiency frontier is moving as fast as the capability frontier. A 3.6x training speedup is the kind of progress that shows up later as a model you can actually run.
← All articles
Category
Ai
327 articles
Ten Reference Images at Once: AI Image Editing Moves from Single Shots to Composites
Single-image editing makes variations. Multi-image editing makes combinations. What ten references at once changes about how we prompt and compose.
Native Transparency Is Quietly the Most Useful New Feature in AI Images
Everyone asks for realism. Designers ask for a transparent background. Why native alpha-channel generation is the quiet feature that changes real workflows.
Alibaba's Qwen-Image 2.1 Just Changed the Open Model Licensing Conversation
The technical specs are impressive, but the license is the real story. Qwen-Image 2.1 drops Apache 2.0 for a research license, and the fine print now matters more than ever.
Tongyi Wanxiang Qwen-Image 2.1 Goes Open Source: How a 7B Small Model Fits Transparent Images and 10-Image Editing on a Single Consumer GPU
A 7B open-source image generation model — how does it squeeze transparent images and multi-image editing onto a 6GB GPU? Breaking down Qwen-Image 2.1's core upgrades and its licensing shift.
Can You Still Tell an AI Image from a Real One? The Honest Answer Is No.
The tells are gone and the detectors are unreliable. The honest answer to spotting AI images is no, and the fix is upstream of your eyeballs.
The Quiet Licensing Shift That Could Change Open-Source AI Images
Open weights no longer mean what they did a year ago. The licensing fine print is getting more complicated right as the models get better.
The Local Model Squeeze: Why Creators Keep Chasing Unrestricted Image Models
Every week another creator asks for an unrestricted model. The demand says more about the cloud tools than the people asking.
Tongyi Wanxiang Qwen-Image 2.1 Goes Open Source: Why a 7B Model Dares to Challenge Google and OpenAI
A 7B open-source image generation model—how can it go head-to-head with Google and OpenAI? This article breaks down Qwen-Image 2.1's scorecard, core capabilities, and licensing controversy.
The Open-Source AI Image Stack Is Being Rebuilt, One Workflow at a Time
Workflow1111 recreates AUTOMATIC1111 with 73 nodes and 11 pipelines. What the community rebuild means for local image generation.
The Numbers Behind AI Image Generation: A 99 Percent Price Drop Ate Stock Photography
Market size, stock photo disruption, adoption numbers and the hybrid workflow, explained through the statistics.
The Next AI Frontier Is Turning a Single Still Image Into a Moving Scene
How the physics problem got solved, the open-source arrivals, and which image-to-video tools are actually usable today.