Model Configuration

Model Size (Parameters)

Custom Size (Billions)

Precision/Quantization

Context Length (tokens)

Batch Size

Primary Use Case

📊 Hardware Requirements

Minimum VRAM Required

Recommended VRAM (with overhead)

System RAM Required

Storage Space

💡 Pro Tips:

This calculator uses the following formulas:

Model Weights: Parameters × Bytes per parameter
KV Cache: 2 × Layers × Hidden size × Context length × Batch size × Precision
Activation Memory: ~10-20% of model weights
Overhead: ~20% for CUDA kernels and temporary buffers