Downloads · 30 days
24
13% of all-time downloads
hibiki-zero-3b-mlx-q4
huybik/hibiki-zero-3b-mlx-q4
hibiki-zero-3b-mlx-q4 is a audio-to-audio model from huybik. Use it for the audio-to-audio task on the model card, and read the license before you ship it in a product. It is set up for mlx. The card lists the license as cc-by-nc-sa-4.0.
4-bit (MLX) quantization of kyutai/hibiki-zero-3b-pytorch-bf16, a simultaneous speech-to-speech + speech-to-text translation model (FR / ES / PT / DE → EN) with voice transfer.
All-time downloads
184
Public
Repo size
5.2 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors2.8 GB · 100%
At a glance
- Task
- Audio-to-Audio
- Library
- mlx
- License
- cc-by-nc-sa-4.0
- Model type
- hibiki
- Access
- Public
- Created
- Jun 5, 2026
- Updated
- Jun 5, 2026
- SHA
- 7704e4f8
Base models
Model card
From the Hugging Face model README
Hibiki-Zero 3B — MLX 4-bit
4-bit (MLX) quantization of
kyutai/hibiki-zero-3b-pytorch-bf16,
a simultaneous speech-to-speech + speech-to-text translation model
(FR / ES / PT / DE → EN) with voice transfer.
The language model is quantized to 4-bit (group_size=32), shrinking it from 5.8 GB → 2.2 GB. On an Apple M4 Pro this runs at ~1.3× real-time (16.5 tok/s @ 12.5 Hz), faster than the bf16 PyTorch/MPS path. The Mimi codec is kept separate in bf16.
group_size=32 is intentional: stock moshi-mlx (and the moshi-swift iOS
loader) hardcode gs32 for .q4.safetensors, so a larger group size would not
load without patching the loader.
Results (Apple M4 Pro, leon.wav, FR→EN)
| Value | |
|---|---|
| LM weights | 5.8 GB bf16 → 2.2 GB q4 (578 layers quantized) |
| Speed | 16.5 tok/s ≈ 1.3× real-time — vs ~0.7× for the PyTorch/MPS path (~1.9× faster) |
| Quality | Coherent FR→EN — correctly translated the Léon Marchand / Paris 2024 Olympics commentary; minor q4 artifacts (e.g. one "Paris 1024" slip) |
| Audio out | 24 kHz wav ✅ |
Files
| File | Description |
|---|---|
hibiki.q4.safetensors | 4-bit quantized LM (MLX) |
config.json | model config (needed to build the LmConfig) |
[email protected] | Mimi codec (bf16) |
tokenizer_spm_48k_multi6_2.model | SentencePiece tokenizer |
mlx_hibiki_patch.py | runtime patches for moshi-mlx (required) |
verify_mlx_q4.py | example inference script |
Why the patch is required
Stock moshi-mlx (0.3.0) targets
moshi / older hibiki and misses three hibiki-zero deltas, so these weights will
not load without mlx_hibiki_patch.py:
- config — honour
hidden_scale(feedforward dim) andkv_repeat=2instead of the hardcoded4*dim/kv_repeat=1. - attention — grouped-query attention in the forward pass (the main
transformer uses
kv_repeat=2). - positional embedding —
rope_concat== RoPE withinterleave=False(MLXtraditional=False).
Usage
pip install moshi-mlx
python - <<'PY'
import mlx_hibiki_patch # patches moshi_mlx for hibiki-zero — import first
from moshi_mlx import run_inference
import sys
sys.argv = [
"run_inference",
"--lm-config", "config.json",
"--moshi-weights", "hibiki.q4.safetensors",
"--mimi-weights", "[email protected]",
"--tokenizer", "tokenizer_spm_48k_multi6_2.model",
"input_fr.wav", "output_en.wav",
]
run_inference.main()
PY
See verify_mlx_q4.py for a ready-to-run example.
License
Inherits CC BY-NC-SA 4.0 (non-commercial, share-alike) from the base model
kyutai/hibiki-zero-3b-pytorch-bf16.
- Task
- Audio-to-Audio
- Library
- mlx
- Type
- hibiki
- License
- cc-by-nc-sa-4.0
- Languages
- en, fr, es, pt, de
- Created
- Jun 5, 2026
- Updated
- Jun 5, 2026