Advanced LLM VRAM Estimator

Precise memory estimation for modern language models

Advanced Options

Weights Memory

0.00 GB

Activations Memory

0.00 GB

KV Cache Memory

0.00 GB

Total Estimated VRAM

0.00 GB
Calculation Details

Weight Memory Calculation:

Parameters (in billions) × Quantization bytes × 1B / 1024³

Activation Memory:

Batch × Context × Hidden Size × Layers × 2 bytes (FP16) / 1024³

KV Cache Memory:

Batch × Seq Length × Layers × (2 × Head Dim × KV Heads) × 2 bytes / 1024³

Hidden Size Estimation:

√(Parameters / (6 × Layers)) or calculated from heads and head dimension

Total VRAM:

(Weights + Activations + KV Cache) × 1.20 (20% overhead)

Note: This is an estimation. Actual VRAM usage depends on implementation, optimizations, and framework overhead.