anerjy
/

Qwen3.5-4B-4bit-text-mlx

4-bit precision

Model card Files Files and versions

Qwen3.5-4B 4bit · text-only (vision tower stripped) · MLX

Derived from mlx-community/Qwen3.5-4B-4bit: vision tensors removed, config flattened for text-only mlx / mlx-swift loading. 924 tensors, 2.37 GB. Made for an offline on-device assistant app. Apache-2.0 (inherits Qwen3.5).

Downloads last month: 15

Safetensors

Model size

0.7B params

Tensor type

BF16

·

U32

·

F32

·

MLX

Hardware compatibility

Log In to add your hardware

4-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for anerjy/Qwen3.5-4B-4bit-text-mlx

Base model

Qwen/Qwen3.5-4B-Base

Finetuned

Qwen/Qwen3.5-4B

Quantized

(332)

this model