WhoBuilt

Key facts

Built by
Capabilities
Architecture was built from scratch rather than using the standard Stable Diffusion architecture
Supports spatial reasoning, including object relationships and novel text
Generates accurate text and supports specifying text placement and size in designs
Supports prompts up to 8,000 tokens

Timeline 5

Dated events, newest first. Open the sources under any item.

How we verify
Built1Capability4
20245 events
  1. as of Sep 2024Capability
    Architecture was built from scratch rather than using the standard Stable Diffusion architecture
    “We built something completely from scratch” 24:07
    Unverified1 source· Talk
    +2 more capabilities on the full timeline
  2. as of Sep 2024Capability
    Supports spatial reasoning, including object relationships and novel text
    “Spatial reasoning is all there, including text that's totally novel” 15:06
    Unverified1 source· Talk
  3. as of Sep 2024Built
    Playground built Playground v3
    “Playground v3 is state of the art in many aspects” 23:56
    Unverified1 source· Talk

Talks 1

55:51
Building The World's Best Image Diffusion Model

Suhail Doshi explains how Playground shifted from a general image generator toward a visual design tool built around templates and conversational editing. The talk describes Playground v3’s focus on prompt understanding, accurate text rendering, and detailed image control, along with the architectural changes the team made to support those goals. Doshi discusses learning from user behavior, choosing commercially useful design tasks, and how close prompt adherence can complicate aesthetic evaluations. He also reflects on earlier ventures, research priorities, and why reaching state of the art requires sustained attention to small details.

Sep 19, 2024· 56 minDiscussed 23:56