Quantization Memory Savings Calculator
Compare model VRAM across fp32, fp16, int8 and int4.
Results update automatically as you type.
Result
int4 needs 42 GB vs 168 GB at fp16
Quantizing a 70B model cuts VRAM up to 75%
| Precision | Bytes/param | VRAM |
|---|---|---|
| fp32 | 4 bytes | 336 GB |
| fp16 | 2 bytes | 168 GB |
| int8 | 1 bytes | 84 GB |
| int4 | 0.5 bytes | 42 GB |
See how quantization shrinks a model's memory footprint. Compare VRAM at float32, float16, int8 and int4 for any parameter count.
How this is calculated
This calculator uses the infrastructure formulas described in our methodology. All figures are estimates for planning; verify current pricing with each provider before relying on them.
Worked example (defaults)
With the default inputs above, here is the result:
Result
int4 needs 42 GB vs 168 GB at fp16
Quantizing a 70B model cuts VRAM up to 75%
| Precision | Bytes/param | VRAM |
|---|---|---|
| fp32 | 4 bytes | 336 GB |
| fp16 | 2 bytes | 168 GB |
| int8 | 1 bytes | 84 GB |
| int4 | 0.5 bytes | 42 GB |
Frequently asked questions
How much does int4 save?+
int4 uses 1/8 the weight memory of float32 and 1/4 of float16 — often the difference between fitting on one GPU or several.
Does quantization reduce quality?+
Modern 8-bit and 4-bit methods keep most quality for many tasks, but always evaluate on your own workload.