Artificial Analysis Ranks xAI Grok Imagine Quality in Top Five

Jun 1, 2026

Artificial Analysis used blind human preference testing to rank grok-imagine-image-quality at #5 on its global image leaderboards. The model achieved an Elo score (a numerical rating used to rank relative model performance) of 1,203, placing it as the highest-rated model outside of the OpenAI and Google ecosystems.

Elo Score: 1,203
Text to Image Rank: 5
Image Editing Rank: 5

The findings highlight a significant shift in the competitive landscape for high-fidelity media. While the model sits behind frontier systems like GPT Image 2, it delivers comparable quality at a much lower cost. Generating 1,000 images at 1K resolution costs $50, which is roughly 75% cheaper than OpenAI’s leading offering.

Artificial Analysis notes that this performance-to-price ratio positions xAI as a viable alternative for professional workflows. The model is available via the Grok app and the Grok Imagine API, supporting 2K resolution and multi-reference image editing. Access is also available through third-party providers like fal for serverless integration.

View the full update on artificialanalysis.ai

Artificial Analysis

@ArtificialAnlys6d ago

grok-imagine-image-quality lands at #5 on both the Artificial Analysis Text to Image and Image Editing leaderboards, the leading model outside of OpenAI and Google and at a much lower price! grok-imagine-image-quality is @xAI's latest image model and a higher quality variant of grok-imagine-image. It sits behind only OpenAI's GPT Image and Google's Nano Banana across both Text to Image and Image Editing. The model supports 2K outputs at $70/1k images and 1K outputs at $50/1k images, with editing supporting up to 3 reference images. At $50/1k images for the 1K resolution tested, grok-imagine-image-quality is cheaper than GPT Image 2 at $211/1k images and Nano Banana Pro at $134/1k images, though priced at a premium compared to the standard grok-imagine-image model at $20/1k images. grok-imagine-image-quality is available via xAI's native API, the Grok chat app, and third party APIs. Congratulations to @xAI and @elonmusk on the launch! See below for comparisons between grok-imagine-image-quality and other leading models in the Artificial Analysis Image Arena 🧵

6111

View on X

Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →

Keep reading

xAI Opens Grok Imagine Quality Mode API for Enterprise Image Generation

xAI released its high-fidelity image generation model to developers, enabling 2K resolution and advanced brand consistency through its public API. The update moves the technology behind Grok's 300 million generated images into a professional tier focused on photorealistic product visualization and marketing assets.

xAI Grok Imagine Video 1.5 Takes Top Spot in Arena Rankings

Arena3d ago

xAI Grok Imagine Video 1.5 Takes Top Spot in Arena Rankings

xAI's Grok-Imagine-Video-1.5-Preview (720p) has reached the #1 position on the Arena Image-to-Video leaderboard with an Elo score of 1,473. The model unseated previous leaders from ByteDance and Alibaba, marking a significant jump in human-preferred video generation quality.

falMay 8

fal Launches Grok Imagine Quality Mode API for 2K Image Generation

fal, a generative media infrastructure platform, now hosts xAI's Grok Imagine Quality Mode API for high-fidelity image generation and editing. This release allows developers to integrate 2K resolution visuals and natural language image manipulation into their applications via serverless infrastructure.

Artificial Analysis ranks HiDream-O1-Image-Dev-2604 as top open weights model

Artificial Analysis2d ago

Artificial Analysis ranks HiDream-O1-Image-Dev-2604 as top open weights model

Artificial Analysis placed HiDream-O1-Image-Dev-2604 at the top of its open-weights image generation leaderboard. The 8B parameter model matches the quality of proprietary systems by using a unified transformer architecture and a dedicated prompt-refinement pipeline.