Glossary / Quantization
QuantizationKuantisasi
Storing a model's numbers with fewer bits so it is smaller and runs on weaker hardware, with a small loss in quality.
Menyimpan angka model dengan bit lebih sedikit supaya lebih kecil dan bisa jalan di perangkat lebih lemah, dengan sedikit penurunan kualitas.
Example
Contoh
Q4 means about 4 bits per weight, Q8 about 8.
Q4 berarti sekitar 4 bit per bobot, Q8 sekitar 8.
Why it matters
Kenapa penting
Q4 is often the best trade-off between size and quality for laptops.
Q4 sering jadi kompromi terbaik antara ukuran dan kualitas untuk laptop.
See it in the calculator →Lihat di kalkulator →