nvidia / automatic-speech-recognition Open weights updated 1 month ago

nemotron-3.5-asr-streaming-0.6b

Name: nemotron-3.5-asr-streaming-0.6b
Author: nvidia

/ Badge alignment consistency / img { display: inline; vertical-align: middle; }

Params

640 M

Context

—

Downloads 30d

560 K

Likes

1,266

Licence: other Not gated GGUF · SAFETENSORS 35 languages View on Hugging Face ↗

Download history

daily snapshots · 106 days

▲ 725 K in the last 30 days (56.4%)

1.3 M560 K

Jul 28Sep 1Oct 6Nov 10

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

File	Quant	Size	Est. VRAM	Verdict on RTX 4090 · 24 GB
nemotron-3.5-asr-streaming-0.6b.q8_0.gguf	Q8_0	0.7 GB	1.4 GB	✅ Runs comfortably
model.safetensors	f32	2.6 GB	3.4 GB	✅ Runs comfortably

Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

Architecture: Nemotron3_5AsrForRNNT
Parameters: 640 M
Tensor type: F32
Vocabulary: 13,088
Licence: other
First seen on the Hub: 2026-05-15
Training datasets: nvidia/Granary, multilingual_librispeech, fleurs, mozilla-foundation/common_voice_8_0, voxpopuli, europarl
FLEURS (Hindi) (reported): 6.81
FLEURS (French) (reported): 9.03
FLEURS (German) (reported): 8.31
FLEURS (Korean) (reported): 7.12
FLEURS (English) (reported): 7.91
FLEURS (Italian) (reported): 4.25
FLEURS (Spanish) (reported): 4.11
FLEURS (Portuguese) (reported): 5.48
Added to our catalog: 2026-07-28

Base model and the most-downloaded derivatives in the catalog.

Compare with any automatic-speech-recognition model

Popular comparisons