Downloads · 30 days
88
34% of all-time downloads
Qwen3.5-2B-OneCompression-4bit-MLX
kizuna-intelligence/Qwen3.5-2B-OneCompression-4bit-MLX
Qwen3.5-2B-OneCompression-4bit-MLX is a text generation model from kizuna-intelligence. Use it when you need the model to write or continue text. It is set up for mlx. The card lists the license as apache-2.0.
Text-only MLX checkpoint produced from Qwen/Qwen3.5-2B for AyaneSDK's on-device conversation example.
All-time downloads
260
Public
Parameters
1.9B
2 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors1 GB · 98%
Parameter types
How the weights are stored.
U321.9B · 100%
Base models
Model card
From the Hugging Face model README
Qwen3.5-2B OneCompression 4-bit MLX
Text-only MLX checkpoint produced from Qwen/Qwen3.5-2B for AyaneSDK's
on-device conversation example.
- OneCompression GPTQ with quantization-error propagation (QEP)
- 4-bit weights, group size 128
- 256 Japanese dialogue calibration samples of 512 tokens
- 186 quantized linear layers
- Token embedding quantized separately to asymmetric MLX 4-bit
- Vision weights are not included
The packed GPTQ-v1 linear weights were converted losslessly to MLX's
row-major affine representation. The model configuration retains an
onecompression_source_quantization audit record.
Source
Base model: Qwen/Qwen3.5-2B
Quantizer: FujitsuResearch/OneCompression
- Task
- Text Generation
- Library
- mlx
- Type
- qwen3_5_text
- License
- apache-2.0
- Languages
- ja, en
- Created
- Sep 2, 2026
- Updated
- Sep 2, 2026