How much VRAM does ultravox-v0_5-llama-3_2-1b need?

Q4_K_M is exactly 807,694,464 bytes (0.75 GiB) in weights. Add the KV cache, which depends on context length, plus roughly half a gigabyte of runtime overhead.

How large is ultravox-v0_5-llama-3_2-1b's KV cache?

8.00 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.

fixie-ai · text

ultravox-v0_5-llama-3_2-1b

fixie-ai/ultravox-v0_5-llama-3_2-1b

ultravox-v0_5-llama-3_2-1b at Q4_K_M is exactly 807,694,464 bytes (0.75 GiB / 0.81 GB) — an effective 9.459 bits per weight, not the nominal 4. Its KV cache at 32K is 8.00 GiB.

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

Parameters

683M

Architecture

llama

16 layers

Context

—

native (config.json)

License

mit

Shipped quantizations

● exact bytes, summed from published files

Quant	Size●	Exact bytes●	Effective bpw●	Tensors●	Publisher
Q4_K_M	0.75 GiB	807,694,464	9.459	—	ggml-org
Q8_0	1.23 GiB	1,321,083,008	15.471	—	ggml-org

KV cache by context

computed per layer

Context	KV cache (f16)●	Flat formula	Overstated by	Full / windowed / recurrent
4,096	1.00 GiB	1.00 GiB	—	16 / 0 / 0
8,192	2.00 GiB	2.00 GiB	—	16 / 0 / 0
16,384	4.00 GiB	4.00 GiB	—	16 / 0 / 0
32,768	8.00 GiB	8.00 GiB	—	16 / 0 / 0
65,536	16.00 GiB	16.00 GiB	—	16 / 0 / 0
131,072	32.00 GiB	32.00 GiB	—	16 / 0 / 0

Compare with

same modality, comparable size

vs Qwen3-4B vs HyperCLOVAX-SEED-Text-Instruct-1.5B vs Llama-3.2-1B-Instruct vs ced-base vs gemma-3-1b-it vs Qwen3-1.7B All alternatives →Will it fit on my card? →

Why other calculators give a different number

A parameters × bits ÷ 8 estimate puts Q4_K_M at roughly 0.36 GiB. The real file is 0.75 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

128

Hidden size

4096

Vocab

128,256

Sliding window

none

SWA period

—

MLA

Experts

—

Experts per token

—

use_sliding_window

—

Questions people ask

How much VRAM does ultravox-v0_5-llama-3_2-1b need?: Q4_K_M is exactly 807,694,464 bytes (0.75 GiB) in weights. Add the KV cache, which depends on your context length, plus roughly half a gigabyte of runtime overhead.
How large is ultravox-v0_5-llama-3_2-1b's KV cache?: 8.00 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 ultravox-v0_5-llama-3_2-1b 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.