Member-only story
Inside Parti, Another Impressive Text-to-Image Model from Google
The new model uses autoregressive methods and transformers to achieve state-of-the-art text-to-image generation with a simpler architecture.
I recently started an AI-focused educational newsletter, that already has over 125,000 subscribers. TheSequence is a no-BS (meaning no hype, no news etc) ML-oriented newsletter that takes 5 minutes to read. The goal is to keep you up to date with machine learning projects, research papers and concepts. Please give it a try by subscribing below:
The Google Brain team seems to be super active in the text-to-image space. Recently, they published details about Imagen, a model that they claim outperforms OpenAI’s famous DALL-E 2. Just a few days after, they published a paper unveiling the Pathways Autoregressive Text-to-Image model (Parti), a simpler text-to-image generation model that supports very rick compositions that involve relevant world knowledge.
