New Logo and Identity for Ideogram by How&How

Prompt-l’œil Prompt and Circumstance Established in 2022, Ideogram (pronounced eye-dee-oh-gram) is an “open visual intelligence model for products, agents, and creative teams”. In other words, a generative AI platform. Co-founded by Mohammad Norouzi, William Chan, Chitwan Saharia, and Jonathan Ho — a team of AI experts who had previously led AI projects at Google Brain, UC Berkeley, CMU, and the University of Toronto — with “a focus on creativity and a high standard for trust and safety”, it launched its first model, 0.1, in 2023 and after multiple iterations and seed rounds along with notable improvements in its ability to generate text and treat it as an editable element, it has recently launched Ideogram 4.0. that, if any of this means anything to you, it is a “9.3B-parameter Diffusion Transformer trained from scratch, paired with a frozen 8B vision-language model as its text encoder. The nf4 checkpoint runs on a single 24GB consumer GPU”. (The amount of times I have put text in quotes should be an indication of how little any of it means to me.) In a blind test of sorts (explained towards the bottom of this link), Ideogram 4.0’s results put it ahead of Nano Banana 2 and behind the leader, OpeanArt’s GPT Image 2. In other words, it’s, I guess, “good”. With the release of the latest model, Ideogram has introduced a new identity designed by London, UK- and Los Angeles, CA-based How&How. /** * Stop all iframes or HTML5 's from playing */ var stopVideos = function () { var videos = document.querySelectorAll('iframe, video'); Array.prototype.forEach.call(videos, function (video) { if (video.tagName.toLowerCase() === 'video') { video.pause(); } else { var src = video.src; video.src = src; } }); }; .pausevideos { margin: 20px 0 30px; display: block; font-size: .8em; font-family: 'ABCSocial'; font-variation-settings: "wdth" 100, "wght" 370, "slnt" 0; -webkit-font-feature-settings: "ss02" 1, "ss04" 1, "ss05" 1, "lnum" 1, "liga" 1; font-feature-