GPU Memory Estimator

Estimate GPU memory requirements for AI models.

Model Configuration

Operation Type

Model Size (Billions of Parameters)

B

Precision

Batch Size

Sequence Length

Total GPU Memory Required

13.6 GB

For inference

🧠Model Weights

13.0 GB

💾KV Cache

0.5 GB

Memory Breakdown

Model Weights13.04 GB

KV Cache0.52 GB

Recommended GPUs

RTX 3090 (24GB)RTX 4080 (16GB)RTX 4090 (24GB)A10 (24GB)A100 40GB (40GB)A100 80GB (80GB)H100 (80GB)