How much VRAM does nemotron-speech-streaming-en-0.6b need?

Q4_K_M is exactly 475,436,032 bytes (0.44 GiB) in weights. Add the KV cache, which depends on context length, plus roughly half a gigabyte of runtime overhead.

nvidia · audio asr

nemotron-speech-streaming-en-0.6b

nvidia/nemotron-speech-streaming-en-0.6b

nemotron-speech-streaming-en-0.6b at Q4_K_M is exactly 475,436,032 bytes (0.44 GiB / 0.48 GB) — an effective 6.154 bits per weight, not the nominal 4.

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

Parameters

618M

Architecture

parakeet

null layers

Context

—

native (config.json)

License

other

Shipped quantizations

● exact bytes, summed from published files

Quant	Size●	Exact bytes●	Effective bpw●	Tensors●	Publisher
Q4_K_M	0.44 GiB	475,436,032	6.154	—	handy-computer
Q5_K_M	0.50 GiB	538,989,568	6.976	—	handy-computer
Q6_K	0.56 GiB	600,420,352	7.771	—	handy-computer
Q8_0	0.68 GiB	729,650,176	9.444	—	handy-computer
F16	1.15 GiB	1,237,652,608	16.019	—	handy-computer
F32	2.30 GiB	2,472,386,176	32.001	—	handy-computer

Measured

published by a third party, attributed below

Metric	Value	What it means
rtf	1071.4
RTFx	1071.4	Higher is better — audio seconds processed per second of compute.
Word error rate	8.84%	Lower is better — the share of words transcribed incorrectly.
Word error rate	12.02%	Lower is better — the share of words transcribed incorrectly.
Word error rate	7.91%	Lower is better — the share of words transcribed incorrectly.
Word error rate	1.88%	Lower is better — the share of words transcribed incorrectly.
Word error rate	4.34%	Lower is better — the share of words transcribed incorrectly.
Word error rate	2.69%	Lower is better — the share of words transcribed incorrectly.
Word error rate	2.40%	Lower is better — the share of words transcribed incorrectly.
Word error rate	5.73%	Lower is better — the share of words transcribed incorrectly.

Benchmarked· by open-asr-leaderboard-english-short-latest

RTFx measured by the Open ASR Leaderboard on a single datacenter GPU at a large batch size. It ranks models against each other; it says nothing about throughput on consumer hardware. We reproduce these figures with attribution; they are not ours and we have not verified the runs. Source: open-asr-leaderboard-english-short-latest.

Compare with

same modality, comparable size

vs nemotron-3.5-asr-streaming-0.6b vs parakeet-unified-en-0.6b vs parakeet-tdt-0.6b-v3 vs whisper-medium vs whisper-large-v3 vs canary-180m-flash 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 0.32 GiB. The real file is 0.44 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

1,025

Sliding window

none

SWA period

—

MLA

Experts

—

Experts per token

—

use_sliding_window

—

Questions people ask

How much VRAM does nemotron-speech-streaming-en-0.6b need?: Q4_K_M is exactly 475,436,032 bytes (0.44 GiB) in weights. Add the KV cache, which depends on your context length, plus roughly half a gigabyte of runtime overhead.
Which quantization of nemotron-speech-streaming-en-0.6b 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.