by Playground AI
SDXL-based model fine-tuned for exceptional aesthetic quality. Consistently ranked top on human preference benchmarks. Excellent at photorealism and artistic compositions. Inherits SDXL ControlNet compatibility — canny, depth, and openpose ControlNets work with varying degrees of success.
VRAM requirements, GPU fit, and setup notes for Playground v2.5, including 8GB/12GB fit guidance where relevant. Recommended runtimes: ComfyUI and Diffusers support. Best download size: ~6.9 GB at FP16.
Your hardware
Detecting...
Measured quality metrics for Playground v2.5 outputs.
How often humans prefer this model's output (0-100%)
Visual quality and composition rating (5-9 scale)
Text-image alignment accuracy (higher is better)
Compare which GPUs can run Playground v2.5 at different precisions. FP8 uses less memory than FP16 when available, and the grade shows how comfortably each GPU handles the workload.
| Resolution | VRAM Required | RTX 4090 24GB | RTX 3060 12GB | RTX 4060 8GB | MacBook Pro M4 Pro 24GB |
|---|---|---|---|---|---|
| 512×512 | 7.7 GB | S● | S● | B● | S● |
| 768×768 | 7.8 GB | S● | S● | B● | S● |
| 1024×1024 | 8.0 GB | S● | S● | B● | S● |
Turbo / LCM distillation
Use distilled scheduler at 4-8 steps for faster iteration
ControlNets available
Add guided generation with 3 adapters (+1.2 GB VRAM each)
from diffusers import StableDiffusionXLPipeline
import torch
pipe = StableDiffusionXLPipeline.from_pretrained(
"playgroundai/playground-v2.5-1024px-aesthetic",
torch_dtype=torch.float16
)
pipe.to("cuda")
image = pipe(
prompt="your prompt here",
num_inference_steps=50,
guidance_scale=3.0,
height=1024,
width=1024,
).images[0]
image.save("output.png")Get started
Setup instructions for running Playground v2.5 locally
1. Download the model
Get the checkpoint from HuggingFace
2. Place in:
ComfyUI/models/checkpoints/3. Launch ComfyUI
python main.pyComfyUI Workflow
Basic txt2img workflow for Playground v2.5
Drag & drop into ComfyUI or use File → Import
VRAM allocation at 1024×1024 on RTX 4090 24GB (24 GB)
Time per image at 1024×1024, 28 steps, FP16.
Download Playground v2.5 in the precision that matches your GPU. Lower precision usually means less VRAM pressure, while higher precision keeps more quality.
| Format | Precision | Size | Provider | |
|---|---|---|---|---|
| safetensorsRecommended | FP16 | 6.9 GB | official | Download |
3 ControlNets available for Playground v2.5. ControlNets add guided image generation (edges, depth, pose) at the cost of extra VRAM.
SDXL canny ControlNet — compatible with Playground v2.5 as an SDXL-based model. Results may vary slightly from base SDXL.
SDXL depth ControlNet — inherited compatibility from SDXL base architecture.
SDXL openpose ControlNet — inherited compatibility from SDXL base architecture.
Some SDXL LoRAs are compatible but results vary.
Frequently asked questions
Playground v2.5 (3.5B parameters) requires approximately 8.0 GB of VRAM at FP16 precision for standard 1024×1024 image generation. If you want a lighter setup, lower precisions like FP8 can reduce memory use when available.
Playground v2.5 usually needs more than 8GB for comfortable local use. Check the VRAM table above for the exact resolution and precision trade-off.
Playground v2.5 is marked for ComfyUI and Diffusers support in our catalog, so those are the runtimes we recommend first for local setup. If your workflow uses another front end, check the model's available formats and workflow notes above before downloading.
Yes, the RTX 4090 (24 GB VRAM) can run Playground v2.5 comfortably at FP16. Expected generation time is around ~6s per image at 1024×1024.
Yes, Playground v2.5 has 3 ControlNet adapters available: Canny Edge (SDXL), Depth Map (SDXL), OpenPose (SDXL). Each ControlNet adds roughly 1.2 GB of extra VRAM.
Some SDXL LoRAs are compatible but results vary. The LoRA ecosystem for Playground v2.5 is rated as "minimal". Each LoRA adds roughly 0.2 GB of extra VRAM.
On a reference GPU (RTX 4090 24GB), Playground v2.5 generates a 1024×1024 image in approximately ~6s at FP16 with 28 inference steps. Faster GPUs with higher memory bandwidth will produce images more quickly.
See also