Voxtral-Mini-3B-2507-TurboQuant-MLX-2bit

2-bit MLX weight-quantized build of mistralai/Voxtral-Mini-3B-2507. Extreme-compression variant with a TurboQuant KV-cache profile — designed for memory-constrained Apple Silicon devices.

Hardware compatibility

Device	VRAM / RAM	Recommendation
Apple M4 Max 128 GB	~1.3 GB	recommended — headroom for long context
Apple M3 Max 64 GB	~1.3 GB	comfortable
Apple M2 Max 32 GB	~1.2 GB	fits

Overview

Base: mistralai/Voxtral-Mini-3B-2507 — 3B speech-understanding model
Capabilities: transcription, speech translation, audio QA
Weight precision: 2-bit (group-wise)
KV-cache profile: TurboQuant (per-head static calibration)
Approx. on-disk size: ~1 GB
Runtime: MLX on Apple Silicon

Expect minor WER degradation vs the 4-bit build. Best used with clean, single-speaker audio.

Quickstart

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("majentik/Voxtral-Mini-3B-2507-TurboQuant-MLX-2bit")

prompt = tokenizer.apply_chat_template(
    [{"role": "user", "content": [{"type": "audio", "path": "sample.wav"},
                                  {"type": "text", "text": "Transcribe this."}]}],
    add_generation_prompt=True,
)
print(generate(model, tokenizer, prompt=prompt, max_tokens=256))

Model specs

Field	Value
Parameters	3B
Weight bits	2
Group size	32
Cache profile	TurboQuant
Size on disk	~1 GB
Target hardware	Apple Silicon (M1/M2/M3/M4)
License	Apache 2.0

RotorQuant vs TurboQuant

	TurboQuant	RotorQuant
Strategy	Per-head static calibration	Rotational online re-basis
Memory reduction	~3.5x on KV-cache	~4x on KV-cache
Best for	Batch transcription	Streaming / code-switching

At 2-bit, RotorQuant often preserves quality better in drifting audio — consider the RotorQuant counterpart for streaming workloads.

Model tree for majentik/Voxtral-Mini-3B-2507-TurboQuant-MLX-2bit

Base model

mistralai/Voxtral-Mini-3B-2507

Quantized

(18)

this model

majentik
/

Voxtral-Mini-3B-2507-TurboQuant-MLX-2bit