Generative AI for images with flawless text — built by the Google Brain team behind Imagen
HQ: Toronto, ON, Canada | Founded: 2022 | Employees: 30–60 (est. 2026) | Stage: Series A ($80M) | Website: https://www.ideogram.ai
Ideogram was founded in 2022 in Toronto by Mohammad Norouzi, William Chan, Chitwan Saharia, and Jonathan Ho — all core contributors to Google Brain's Imagen and leading text-to-image research. The company stayed in research through 2022-early 2023, then launched ideogram.ai in August 2023 with a $16.5M CAD (~$22.3M CAD per BetaKit including conversion nuance) seed led by Andreessen Horowitz and Index Ventures. Its thesis: reliable typography inside generated images unlocks marketing, design and logo use cases where AI images previously failed. Ideogram 1.0, 2.0 and Canvas releases emphasize design tools (magic fill, expand, Canvas for layout) beyond prompt→image. In February 2024 it announced an $80M Series A led by a16z with participation from Index, Accel, Radu Timofte, Geoffrey Hinton, and others; valuation was not disclosed but secondary markets priced it around $400-545M.
Ideogram is the typography fix for generative images: four ex-Google Brain researchers — Mohammad Norouzi (CEO) plus William Chan, Chitwan Saharia, Jonathan Ho — left to solve the one thing Midjourney and DALL·E failed at: legible text. Founded 2022, it raised $16.5M CAD at launch (Aug 2023) led by a16z and Index, then $80M Series A Feb 2024 led by a16z at an undisclosed ~$400-500M valuation, backed by AI legends. Its 2.0 and Canvas models are design-workflow native. The moat is talent density and text rendering quality; the risk is that Foundation model scale (Midjourney, Flux, Google, OpenAI) commoditizes image generation and squeezes standalone pricing power.
Browse: Home | Blog | About | All Startups A-Z | View full profile on StartupWiki