CoolFace
Modelpublic

Kelnux/Qwen3-0.6B-seo-finetuned-GGUF

sourceHugging Faceapache-2.0updated 8mo agoView on Hugging Face
0likes39downloads
Model Card

Qwen3-0.6B SEO Fine-tuned (GGUF)

GGUF versions of Kelnux/Qwen3-0.6B-seo-finetuned for use with llama.cpp, Ollama, LM Studio, etc.

Available Files

FileQuantizationSize
Qwen3-0.6B-seo-finetuned-f16.ggufF16 (full precision)~1.1 GB
Qwen3-0.6B-seo-finetuned-q8_0.ggufQ8_0 (8-bit quantized)~610 MB

Training Details

  • —Base model: Qwen/Qwen3-0.6B
  • —Method: LoRA (r=16, alpha=32)
  • —Dataset: metehan777/global-seo-knowledge (2,065 examples)
  • —Final loss: 1.14
  • —Token accuracy: 79.8%

Usage

With llama.cpp

bash
llama-cli -m Qwen3-0.6B-seo-finetuned-q8_0.gguf -p "What is robots.txt in SEO?" -n 200

With Ollama

FROM ./Qwen3-0.6B-seo-finetuned-q8_0.gguf

With LM Studio

Download the GGUF file and load it directly in LM Studio.