TL;DRSpaceXAI shipped Grok Imagine Image 2.0 with editing-first features: magic wand, segmentation, background removal, multi-reference inputs, templates. Ranks second on Arena behind OpenAI. Available on grok.com/imagine. API coming soon.
SpaceXAI shipped Grok Imagine Image 2.0 on Thursday, making editing a first-class capability rather than an afterthought. The update adds a magic wand tool that changes only the region a user points at, segmentation for selecting precise areas to modify, and background removal that exports subjects with transparency. Multi-reference editing accepts up to five input images in a single generation, removing the manual compositing step. A smart-resize feature recomposes one image across nine aspect ratios, from tall banners to wide formats, with the model filling in the new frame.
The model claims the number two spot on both the Arena text-to-image and image-edit leaderboards, behind OpenAI’s gpt-image-2. SpaceXAI launched Grok 4.5 as its first coding model with Cursor last week, and Image 2.0 extends the push into creative tools. The company says it trained the model for fidelity across photography, design, and illustration, with an emphasis on typography and layout so dense visuals hold together and small text renders sharply. Templates for product shots, headshots, e-commerce listings, game assets, and marketing posters package common jobs into ready-made starting points.
