Skip to content
ModelListedExternal
Download

UVOX 96k Llama 32 3B Instruct

Test ultravox model. More coming soon... I hope so.

KindOtherVersionRevision 45c149aCgrade

Model source

Kind
Other
File size
94.5 MB
Version
vRevision 45c149a
Updated
Oct 25, 2026
Downloads
17
Source
Hugging Face

Source description

Test ultravox model. More coming soon... I hope so.

import transformers
import numpy as np
import librosa

pipe = transformers.pipeline(model='AtAndDev/UVOX-96k-Llama-3.2-3B-Instruct', trust_remote_code=True, device="cuda")

path = "voice_input.mp3"
audio, sr = librosa.load(path, sr=16000)

turns = []
pipe({'audio': audio, 'turns': turns, 'sampling_rate': sr}, max_new_tokens=100)

Sources

1 source
Hugging FacePrimaryVersion 45c149a0bb52b8bc52ac07394f9da26b5bbaf9fd
huggingface.co/AtAndDev/UVOX-96k-Llama-32-3B-Instruct/tree/45c149a0bb52b8bc52ac07394f9da26b5bbaf9fd

Verified Oct 25

Model artifacts

1 artifact

model.safetensors

safetensors · 94.5 MB · SHA-256 521716710ac8…25a2 · Hugging Face

Download file

Source excerpts

3 excerpts

--- language: - ar - be - bg - bn - cs - cy - da - de - el - en - es - et - fa - fi - fr - gl - hi - hu - it - ja - ka - lt - lv - mk - mr - nl - pl - pt - ro - ru - sk - sl - sr - sv - sw - ta - th - tr - uk - ur - vi - zh library_name: transformers license: mit metrics: - bleu pipeline_tag: audio-text-to-text --- Test ultravox model. More coming soon... I hope so. ```python import transformers import numpy as np import librosa pipe = transformers.pipeline(model='AtAndDev/UVOX-96k-Llama-3.2-3B-Instruct', trust_remote_code=True, device="cuda") path = "voice_input.mp3" audio, sr = librosa.load(path, sr=16000) turns = [] pipe({'audio': audio, 'turns': turns, 'sampling_rate': sr}, max_new_tokens=100) ```

Source context: 17 downloads · 0 likes · Pipeline audio-text-to-text · Library transformers · Repo AtAndDev/UVOX-96k-Llama-32-3B-Instruct

Source context: 6 downloads · 0 likes · Pipeline audio-text-to-text · Library transformers · Repo AtAndDev/UVOX-96k-Llama-32-3B-Instruct