AI Model Performance Analysis for Text-to-Image Generation

We have updated the Artificial Analysis Text to Image Arena to expand the range of use cases and capabilities we test for, from UI/UX Design to Live-Action Film, and from Physics to Text Rendering. Our updated evaluation measures not just which model is best overall, but which model is best for your specific work. The new leaderboard is live and voting is open. Text to image models have grown more capable over the past years, evolving from single-subject renders in narrow domains, to rich multi-element scenes adhering to fine-grained user control, to systems that render precise typography and interleaved text and graphic layouts. The model an agency uses for campaign creative is not necessarily the model a product manager uses for UI mockups, or a game studio for concept art. Choosing a model today means trading off quality across the specific work you do, and our refreshed methodology answers exactly that question. Our updated evaluation measures against a taxonomy of 10 real-world use cases, 9 model capabilities, and a wide range of styles and aesthetics. We test for use cases such as Marketing & Advertising, Retail & Ecommerce, Live-Action Film, Animation & Gaming, Architecture & Real Estate, Productivity & Knowledge Work, UIUX Design, and capabilities such as Reasoning, Knowledge, Text Rendering, Layout, Physics, and Human Anatomy. Our overall quality benchmark now samples evenly across use cases and capabilities. Each prompt is human curated from anonymized crowdsourced data, and refreshed monthly. Prompts that stop separating models or drift from real world usage get retired, and new ones are added at the capability frontier, so they track the frontier and cannot be overfit. Initial insights from an in-depth analysis of the 10 highest ranking models on the Artificial Analysis Text to Image Leaderboard: ➤ GPT Image 2 is #1 on every use case and capability leaderboard, but the picture changes in the lower price tiers. ➤ Nano Banana 2 is the value pick almost everywhere, ranking top 3 in Animation & Gaming, Consumer, Marketing & Advertising, and UI/UX at $67 per 1,000 images against GPT Image 2's $211. ➤ MAI-Image-2.5 is the best cheap option for Retail & Ecommerce and is strong on Layout. ➤ Nano Banana Pro is the specialist for Live-Action Film, and the cheapest model ranking right after GPT Image 2 (#1), and it also does well on Knowledge and Human Anatomy. See below for the use case and capability breakdowns 🧵