chore(card): add hardware compatibility section
Browse files
README.md
CHANGED
|
@@ -4,24 +4,30 @@ library_name: mlx
|
|
| 4 |
license: apache-2.0
|
| 5 |
pipeline_tag: automatic-speech-recognition
|
| 6 |
tags:
|
| 7 |
-
- voxtral
|
| 8 |
-
- audio
|
| 9 |
-
- speech
|
| 10 |
-
- speech-recognition
|
| 11 |
-
- transcription
|
| 12 |
-
- translation
|
| 13 |
-
- mlx
|
| 14 |
-
- rotorquant
|
| 15 |
-
- quantization
|
| 16 |
-
- 2-bit
|
| 17 |
-
language:
|
| 18 |
-
- en
|
| 19 |
---
|
| 20 |
|
| 21 |
# Voxtral-Mini-3B-2507-RotorQuant-MLX-2bit
|
| 22 |
|
| 23 |
2-bit MLX weight-quantized build of [`mistralai/Voxtral-Mini-3B-2507`](https://huggingface.co/mistralai/Voxtral-Mini-3B-2507) with a RotorQuant KV-cache profile. Ultra-compact, best-available 2-bit stability for streaming audio on Apple Silicon.
|
| 24 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 25 |
## Overview
|
| 26 |
|
| 27 |
- **Base:** `mistralai/Voxtral-Mini-3B-2507` — 3B speech-understanding model
|
|
|
|
| 4 |
license: apache-2.0
|
| 5 |
pipeline_tag: automatic-speech-recognition
|
| 6 |
tags:
|
| 7 |
+
- voxtral
|
| 8 |
+
- audio
|
| 9 |
+
- speech
|
| 10 |
+
- speech-recognition
|
| 11 |
+
- transcription
|
| 12 |
+
- translation
|
| 13 |
+
- mlx
|
| 14 |
+
- rotorquant
|
| 15 |
+
- quantization
|
| 16 |
+
- 2-bit
|
|
|
|
|
|
|
| 17 |
---
|
| 18 |
|
| 19 |
# Voxtral-Mini-3B-2507-RotorQuant-MLX-2bit
|
| 20 |
|
| 21 |
2-bit MLX weight-quantized build of [`mistralai/Voxtral-Mini-3B-2507`](https://huggingface.co/mistralai/Voxtral-Mini-3B-2507) with a RotorQuant KV-cache profile. Ultra-compact, best-available 2-bit stability for streaming audio on Apple Silicon.
|
| 22 |
|
| 23 |
+
## Hardware compatibility
|
| 24 |
+
|
| 25 |
+
| Device | VRAM / RAM | Recommendation |
|
| 26 |
+
| --- | --- | --- |
|
| 27 |
+
| Apple M4 Max 128 GB | ~1.3 GB | recommended — headroom for long context |
|
| 28 |
+
| Apple M3 Max 64 GB | ~1.3 GB | comfortable |
|
| 29 |
+
| Apple M2 Max 32 GB | ~1.2 GB | fits |
|
| 30 |
+
|
| 31 |
## Overview
|
| 32 |
|
| 33 |
- **Base:** `mistralai/Voxtral-Mini-3B-2507` — 3B speech-understanding model
|