Qwen3-4B-SFT-UltraChat-GGUF

GGUF quantized versions of ermiaazarkhalili/Qwen3-4B-SFT-UltraChat for use with llama.cpp, Ollama, LM Studio, and other GGUF-compatible tools.

Available Quantizations

File	Quantization	Quality	Use Case
qwen3-4b-sft-ultrachat-q4_k_m.gguf	Q4_K_M	Good	Recommended - Best balance of quality and size
qwen3-4b-sft-ultrachat-q5_k_m.gguf	Q5_K_M	Better	Higher quality, moderate size increase
qwen3-4b-sft-ultrachat-q8_0.gguf	Q8_0	Best	Highest quality quantization

Download Specific Quantization

Using huggingface-cli

# Download Q4_K_M (recommended)
huggingface-cli download ermiaazarkhalili/Qwen3-4B-SFT-UltraChat-GGUF qwen3-4b-sft-ultrachat-q4_k_m.gguf --local-dir ./models

# Download Q5_K_M (higher quality)
huggingface-cli download ermiaazarkhalili/Qwen3-4B-SFT-UltraChat-GGUF qwen3-4b-sft-ultrachat-q5_k_m.gguf --local-dir ./models

# Download Q8_0 (best quality)
huggingface-cli download ermiaazarkhalili/Qwen3-4B-SFT-UltraChat-GGUF qwen3-4b-sft-ultrachat-q8_0.gguf --local-dir ./models

# Download all quantizations
huggingface-cli download ermiaazarkhalili/Qwen3-4B-SFT-UltraChat-GGUF --local-dir ./models

Using wget

# Q4_K_M
wget https://huggingface.co/ermiaazarkhalili/Qwen3-4B-SFT-UltraChat-GGUF/resolve/main/qwen3-4b-sft-ultrachat-q4_k_m.gguf

# Q5_K_M
wget https://huggingface.co/ermiaazarkhalili/Qwen3-4B-SFT-UltraChat-GGUF/resolve/main/qwen3-4b-sft-ultrachat-q5_k_m.gguf

# Q8_0
wget https://huggingface.co/ermiaazarkhalili/Qwen3-4B-SFT-UltraChat-GGUF/resolve/main/qwen3-4b-sft-ultrachat-q8_0.gguf

Usage

Ollama

# Pull specific quantization
ollama pull hf.co/ermiaazarkhalili/Qwen3-4B-SFT-UltraChat-GGUF:Q4_K_M

# Or create from local file
cat > Modelfile << EOF
FROM ./qwen3-4b-sft-ultrachat-q4_k_m.gguf
EOF

ollama create qwen3-4b-sft-ultrachat -f Modelfile
ollama run qwen3-4b-sft-ultrachat

llama.cpp

# Run with llama-cli
./llama-cli -m qwen3-4b-sft-ultrachat-q4_k_m.gguf -p "Your prompt here" -n 256

# Run as server
./llama-server -m qwen3-4b-sft-ultrachat-q4_k_m.gguf --host 0.0.0.0 --port 8080

llama-cpp-python

from llama_cpp import Llama

llm = Llama(
    model_path="qwen3-4b-sft-ultrachat-q4_k_m.gguf",
    n_ctx=2048,
    n_gpu_layers=-1  # Use all GPU layers
)

output = llm(
    "What is machine learning?",
    max_tokens=256,
    temperature=0.7,
)
print(output['choices'][0]['text'])

LM Studio

1. Download the desired GGUF file from this repository
2. Open LM Studio and navigate to the Models tab
3. Click "Add Model" and select the downloaded GGUF file
4. Load the model and start chatting

GPT4All

1. Download the Q4_K_M GGUF file
2. Open GPT4All and go to Settings > Models
3. Add the GGUF file path
4. Select the model and start using

Original Model

This is a quantized version of ermiaazarkhalili/Qwen3-4B-SFT-UltraChat. See the original model card for:

• Training details and methodology
• Dataset information
• Performance metrics
• Full usage examples with Transformers

Conversion Details

Property	Value
Source Model	ermiaazarkhalili/Qwen3-4B-SFT-UltraChat
Conversion Date	2025-12-26
Quantizations	Q4_K_M, Q5_K_M, Q8_0
Converter	llama.cpp

License

Same license as the original model. See ermiaazarkhalili/Qwen3-4B-SFT-UltraChat for details.

---

Converted using the Slurm Model Trainer skill