alwoolley
/

Qwen3-4B-bnb-8bit

8-bit precision

Model card Files Files and versions

Qwen3-4B-bnb-8bit

Quantized 8-bit BitsAndBytes version of Qwen/Qwen3-4B with 16-bit outlier preservation (threshold=6.0).

Downloads last month: -

Safetensors

Model size

4B params

Tensor type

BF16

·

Inference Providers NEW

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

Model tree for alwoolley/Qwen3-4B-bnb-8bit

Base model

Qwen/Qwen3-4B-Base

Finetuned

Quantized

(311)

this model