Voxtral-4B-TTS-2603-RotorQuant-MLX-2bit

2-bit MLX weight-quantized build of mistralai/Voxtral-4B-TTS-2603 with a RotorQuant KV-cache profile. Ultra-compact multilingual TTS for memory-constrained Apple Silicon; RotorQuant's rotational re-basis provides the best available 2-bit stability across voices and languages.

Overview

Base: mistralai/Voxtral-4B-TTS-2603 — 4B multilingual TTS with zero-shot voice cloning
Weight precision: 2-bit (group-wise)
KV-cache profile: RotorQuant
Approx. on-disk size: ~1.2 GB
Runtime: MLX on Apple Silicon

Quickstart

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("majentik/Voxtral-4B-TTS-2603-RotorQuant-MLX-2bit")

prompt = tokenizer.apply_chat_template(
    [{"role": "user", "content": [
        {"type": "audio", "path": "reference_voice.wav"},
        {"type": "text", "text": "Short prompt."},
    ]}],
    add_generation_prompt=True,
)
audio_tokens = generate(model, tokenizer, prompt=prompt, max_tokens=1024)

Model specs

Field	Value
Parameters	4B
Weight bits	2
Group size	32
Cache profile	RotorQuant
Languages	9
Voice cloning	Zero-shot
Size on disk	~1.2 GB
Target hardware	Apple Silicon (M1/M2/M3/M4)
License	Apache 2.0

RotorQuant vs TurboQuant

	RotorQuant	TurboQuant
Strategy	Rotational online re-basis	Per-head static calibration
Memory reduction	~4x on KV-cache	~3.5x on KV-cache
Best for	Multi-voice / multi-language batches	Single-voice sessions

Model tree for majentik/Voxtral-4B-TTS-2603-RotorQuant-MLX-2bit

Base model

mistralai/Ministral-3-3B-Base-2512

Finetuned

mistralai/Voxtral-4B-TTS-2603

Quantized

(10)

this model