File size: 424 Bytes
fc5280f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
---
base_model: Qwen/Qwen3.5-9B
library_name: mlx
tags:
- mlx
- mlx-lm
- qwen3.5
- osirisbrain
---
# OsirisCortex-v7c MLX Q4

OsirisBrain sovereign reasoning model. Qwen3.5-9B quantized to 4-bit for Apple Silicon Metal GPU.

- **Base model:** Qwen/Qwen3.5-9B
- **Quantization:** 4-bit (group_size=64, affine)
- **Framework:** MLX (Apple Silicon native)
- **Use case:** Core reasoning brain for OsirisBrain AGI Desktop Agent