ComfyUI-SUPIR on 4GB GPU

kijai image_generationtext 3072MB VRAM

Compatible with 4GB GPU

This node will run on RTX 3050, GTX 1650 and similar GPUs with 4GB VRAM.

Optimized Settings for 4GB

VRAM Utilization

75%

You're using 75% of your 4GB VRAM, leaving 1.0GB headroom.

Recommended Batch Size

1

Start with batch size of 1 for optimal performance.

Suggested Resolution

512x512

Best balance of quality and performance for 4GB.

Performance Tips

  • Close other GPU-intensive applications
  • Limited headroom - reduce batch size if OOM occurs
  • Low VRAM mode available
  • Monitor GPU usage with nvidia-smi

🚀 Deploy ComfyUI on Vast.ai

Need low-cost, high-performance cloud GPUs to run ComfyUI and this custom node? Rent instantly on Vast.ai starting at pennies per hour.

Deploy on Vast.ai

🚀 Deploy ComfyUI on RunPod

Need high-performance cloud GPUs to run ComfyUI and this custom node? Rent instantly on RunPod starting at just a few cents per hour.

Deploy on RunPod

VRAM Requirements

Base VRAM Requirement:
3072MB (3.00GB)
Your GPU (4GB):
4096MB (4GB) - RTX 3050, GTX 1650
VRAM Utilization:
75%
Available Headroom:
1.00GB
Low VRAM Mode:
✓ Supported
Estimation Confidence:
MEDIUM

Python Dependencies

  • accelerate
  • omegaconf
  • open-clip-torch>=2.24.0
  • pillow>=9.4.0
  • pytorch-lightning>=2.5.5
  • transformers>=4.28.1

View for Different GPU