Training Design for Text-to-Image Models: Lessons from Ablations

Training Design for Text-to-Image Models: Lessons from Ablations Training Design for Text-to-Image Models: Lessons from Ablations Team Article Published February 3, 2026 Upvote 77 David Bertoin Bertoin Photoroom Roman Frigg photoroman Photoroom Jon Almazán jon-almazan Photoroom Welcome back! This is the second part of our series on training efficient text-to-image models from scratch .

This Research is relevant to the technology intelligence record because it involves Cohere. The source article should remain the factual reference for follow-up coverage.

• Training Design for Text-to-Image Models: Lessons from Ablations Team Article Published February 3, 2026 Upvote 77 David Bertoin Bertoin Photoroom Roman Frigg photoroman Photoroom Jon Almazán jon-almazan Photoroom Welcome back!
• This is the second part of our series on training efficient text-to-image models from scratch .
• In the first post of this series , we introduced our goal: training a competitive text-to-image foundation model entirely from scratch, in the open, and at scale.
• We focused primarily on architectural choices and motivated the core design decisions behind our model PRX .
• We also released an early, small (1.2B parameters) version of the model as a preview of what we are building (go try it if you haven't already 😉).
• In this post, we shift our focus from architecture to training.