Skip to main content
ModelComputer Visionv2.5

PlaygroundAI v2.5

by Playground AI · open-source · Last verified 2026-03-17

PlaygroundAI v2.5 is an open-source text-to-image model that achieved top scores on human preference benchmarks at the time of its release, surpassing SDXL and DALL-E 3 in aesthetic quality evaluations. Built by Playground AI, it uses a novel training approach focused on human aesthetic preference rather than just perceptual similarity metrics.

https://playground.com
C+
C+Average
Adoption: C+Quality: AFreshness: B+Citations: BEngagement: F

Specifications

License
Playground v2.5 Community License
Pricing
open-source
Capabilities
text-to-image, high-aesthetic-quality, fine-tuning, 1024px-output, human-preference-alignment
Integrations
huggingface, diffusers, comfyui
Use Cases
consumer-art, social-media-content, aesthetic-exploration, research, product-imagery
API Available
Yes
Parameters
~2.6B
Context Window
N/A
Modalities
text, image
Training Cutoff
2023
Tags
image-generation, text-to-image, open-source, aesthetic, consumer
Added
2026-03-17
Completeness
100%

Index Score

52.8
Adoption
52
Quality
85
Freshness
70
Citations
60
Engagement
0

Explore the full AI ecosystem on Agents as a Service