jc-builds/SmolLM2-360M-Instruct-Q4_K_M-GGUF

SmolLM2-360M-Instruct Q4_K_M GGUF

This is a Q4_K_M quantized GGUF conversion of HuggingFaceTB/SmolLM2-360M-Instruct optimized for on-device inference with llama.cpp.

Model Details

Property	Value
Original Model	SmolLM2-360M-Instruct
Parameters	360 million
Quantization	Q4_K_M (4-bit, medium quality)
File Size	~258 MB
Context Window	8,192 tokens
Architecture	LLaMA-style transformer
Training Data	4 trillion tokens

Intended Use

This model is optimized for:

• Mobile/Edge Deployment: Runs efficiently on all iOS devices
• llama.cpp Integration: Compatible with llama.cpp and its bindings
• On-Device AI: Private, offline inference without cloud dependencies

Capabilities

• Fast Responses: Very quick inference
• Text Rewriting: Good at summarization and rewriting
• Instruction Following: Solid instruction following for its size
• Low Resource Usage: Minimal RAM and storage requirements
• Trained on 4T Tokens: Quality upgrade from 135M version

Usage with llama.cpp

./llama-cli -m SmolLM2-360M-Instruct.Q4_K_M.gguf -p "Your prompt here" -n 512

License

This model inherits the Apache 2.0 license from the original SmolLM2 model.

Attribution

• Original Model: SmolLM2-360M-Instruct by Hugging Face
• Quantization: jc-builds