Ideogram AI Unveils Photorealistic Text-to-Image Model, Raises $80M to Advance Generative AI

Ideogram AI, founded by former Google engineers, has released Ideogram 1.0, claimed to be the most advanced text-to-image model yet in terms of photorealism, prompt adherence, and text rendering.

The company raised $80 million in Series A funding led by Andreessen Horowitz to further develop its generative AI technology.

Ideogram 1.0 outperforms competitors like Dall-E 3 and MidJourney in side-by-side testing, with superior spatial understanding and ability to generate coherent text.

The model has more customizable censorship compared to others, allowing some images of celebrities and brands that Dall-E 3 and MidJourney block.

With a free tier and paid subscriptions, Ideogram currently seems to be the most capable publicly available image generator, though models like Stable Diffusion 3 may challenge it soon.