How much VRAM does Kokoro-82M need?

Q8_0 is exactly 141,322,336 bytes (0.13 GiB) in weights. Add the KV cache, which depends on context length, plus roughly half a gigabyte of runtime overhead.

hexgrad · audio tts

Kokoro-82M

hexgrad/Kokoro-82M

Kokoro-82M at Q8_0 is exactly 141,322,336 bytes (0.13 GiB / 0.14 GB) — an effective 13.833 bits per weight, not the nominal 8.

From the file· summed from 1 file(s)From the file· KV per layer

Parameters

82M

Architecture

kokoro

3 layers

Context

—

native (config.json)

License

apache-2.0

Shipped quantizations

● exact bytes, summed from published files

Quant	Size●	Exact bytes●	Effective bpw●	Tensors●	Publisher
Q8_0	0.13 GiB	141,322,336	13.833	—	cstr
Q8_0	0.13 GiB	141,322,752	13.833	—	simonfxr
F16	0.15 GiB	163,728,096	16.026	—	cstr
F16	0.15 GiB	163,728,096	16.026	—	simonfxr

Compare with

same modality, comparable size

vs Ace-Step1.5 vs VieNeu-TTS-0.3B vs chatterbox All alternatives →Will it fit on my card? →

Why other calculators give a different number

A parameters × bits ÷ 8 estimate puts Q8_0 at roughly 0.04 GiB. The real file is 0.13 GiB, because a quantization is a mixture and some tensors are always kept at higher precision.

Architecture

from config.json

Layers

Attention heads

—

KV heads

—

Head dim

—

Hidden size

—

Vocab

—

Sliding window

none

SWA period

—

MLA

Experts

—

Experts per token

—

use_sliding_window

—

Questions people ask

How much VRAM does Kokoro-82M need?: Q8_0 is exactly 141,322,336 bytes (0.13 GiB) in weights. Add the KV cache, which depends on your context length, plus roughly half a gigabyte of runtime overhead.
Which quantization of Kokoro-82M should I use?: Q4_K_M is the usual default. Pick the largest quantization that fits your card at the context you actually need — the table above gives exact sizes for every one published.