finnianx/GLM4.6-Qwen3-4b-Instruct-Distill

Model Overview:

A distilled 4B parameter instruction model, designed to be a lightweight and efficient conversationalist.

Purpose:

Follows the style and formatting of GLM4.6, providing much more concise answers than the base Qwen model.

Base Architecture:

Qwen/Qwen3-4B-Instruct-2507

Distillation Source:

The capabilities and response style are derived from GLM-4.6.

Training Dataset:

Fine-tuned on finnianx/glm4.6-2.16k.

Edit:

I dont like this model that much anymore, please try any other model from me or someone else, dont use this!!!