theprint/TiTan-Llama-3.2-1B-GGUF

TiTan-Llama-3.2-1B - GGUF Quantized

Quantized GGUF versions of TiTan-Llama-3.2-1B for use with llama.cpp and other GGUF-compatible inference engines.

Original Model

• Base model: unsloth/Llama-3.2-1B
• Fine-tuned model: theprint/TiTan-Llama-3.2-1B
• Quantized by: theprint

Available Quantizations

• TiTan-Llama-3.2-1B-f16.gguf (2364.7 MB) - 16-bit float (original precision, largest file)
• TiTan-Llama-3.2-1B-q3_k_m.gguf (658.8 MB) - 3-bit quantization (medium quality)
• TiTan-Llama-3.2-1B-q4_k_m.gguf (770.3 MB) - 4-bit quantization (medium, recommended for most use cases)
• TiTan-Llama-3.2-1B-q5_k_m.gguf (869.3 MB) - 5-bit quantization (medium, good quality)
• TiTan-Llama-3.2-1B-q6_k.gguf (974.5 MB) - 6-bit quantization (high quality)
• TiTan-Llama-3.2-1B-q8_0.gguf (1259.9 MB) - 8-bit quantization (very high quality)

Usage

With llama.cpp

# Download recommended quantization
wget https://huggingface.co/theprint/TiTan-Llama-3.2-1B-GGUF/resolve/main/TiTan-Llama-3.2-1B-q4_k_m.gguf

# Run inference
./llama.cpp/main -m TiTan-Llama-3.2-1B-q4_k_m.gguf \
  -p "Your prompt here" \
  -n 256 \
  --temp 0.7 \
  --top-p 0.9

With other GGUF tools

These files are compatible with:

• llama.cpp
• Ollama (import as custom model)
• KoboldCpp
• text-generation-webui

Quantization Info

Recommended: q4_k_m provides the best balance of size, speed, and quality for most use cases.

For maximum quality: Use q8_0 or f16
For maximum speed/smallest size: Use q3_k_m or q4_k_s

License

apache-2.0

Citation

@misc{titan_llama_3.2_1b_gguf,
  title={TiTan-Llama-3.2-1B GGUF Quantized Models},
  author={theprint},
  year={2025},
  publisher={Hugging Face},
  url={https://huggingface.co/theprint/TiTan-Llama-3.2-1B-GGUF}
}