How much VRAM does csm-1b need?

Q4_K is exactly 1,435,926,432 bytes (1.34 GiB) in weights. Add the KV cache, which depends on context length, plus roughly half a gigabyte of runtime overhead.

How large is csm-1b's KV cache?

1.00 GiB at 32K context with an f16 cache, computed per layer from the model's architecture. Quantizing the cache to q8_0 roughly halves it.

sesame · audio tts

csm-1b

sesame/csm-1b

csm-1b at Q4_K is exactly 1,435,926,432 bytes (1.34 GiB / 1.44 GB) — an effective 7.398 bits per weight, not the nominal 4. Its KV cache at 32K is 1.00 GiB.

From the file· summed from 1 file(s)From the file· KV from mirror (mirror:unsloth/csm-1b)

Parameters

1.6B

Architecture

csm

16 layers

Context

2,048

native (config.json)

License

Shipped quantizations

● exact bytes, summed from published files

Quant	Size●	Exact bytes●	Effective bpw●	Tensors●	Publisher
Q4_K	1.34 GiB	1,435,926,432	7.398	—	cstr
Q8_0	1.87 GiB	2,006,354,848	10.337	—	cstr
F16	3.33 GiB	3,574,186,464	18.414	—	cstr

KV cache by context

computed per layer

Context	KV cache (f16)●	Flat formula	Overstated by	Full / windowed / recurrent
4,096	0.13 GiB	0.13 GiB	—	16 / 0 / 0
8,192	0.25 GiB	0.25 GiB	—	16 / 0 / 0
16,384	0.50 GiB	0.50 GiB	—	16 / 0 / 0
32,768	1.00 GiB	1.00 GiB	—	16 / 0 / 0
65,536	2.00 GiB	2.00 GiB	—	16 / 0 / 0
131,072	4.00 GiB	4.00 GiB	—	16 / 0 / 0

Compare with

same modality, comparable size

vs Ace-Step1.5 vs Qwen3-TTS-12Hz-0.6B-Base vs OmniVoice vs orpheus-3b-0.1-ft vs neutts-air vs s2-pro All alternatives →Will it fit on my card? →

Why other calculators give a different number

A parameters × bits ÷ 8 estimate puts Q4_K at roughly 0.81 GiB. The real file is 1.34 GiB, because a quantization is a mixture and some tensors are always kept at higher precision.

Architecture

from mirror:unsloth/csm-1b

Layers

Attention heads

KV heads

Head dim

Hidden size

2048

Vocab

2,051

Sliding window

none

SWA period

—

MLA

Experts

—

Experts per token

—

use_sliding_window

—

Questions people ask

How much VRAM does csm-1b need?: Q4_K is exactly 1,435,926,432 bytes (1.34 GiB) in weights. Add the KV cache, which depends on your context length, plus roughly half a gigabyte of runtime overhead.
How large is csm-1b's KV cache?: 1.00 GiB at 32K context with an f16 cache, computed per layer. Quantizing the cache to q8_0 roughly halves it, which is often the difference between a context length fitting and not.
Which quantization of csm-1b 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.