Instructions to use kevinharry/SmolLM2-135M-4-bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use kevinharry/SmolLM2-135M-4-bit with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="kevinharry/SmolLM2-135M-4-bit",
	filename="SmolLM2-135M_q4_0_quantized.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use kevinharry/SmolLM2-135M-4-bit with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf kevinharry/SmolLM2-135M-4-bit:Q4_0_QUANTIZED
# Run inference directly in the terminal:
llama-cli -hf kevinharry/SmolLM2-135M-4-bit:Q4_0_QUANTIZED

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf kevinharry/SmolLM2-135M-4-bit:Q4_0_QUANTIZED
# Run inference directly in the terminal:
llama-cli -hf kevinharry/SmolLM2-135M-4-bit:Q4_0_QUANTIZED

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf kevinharry/SmolLM2-135M-4-bit:Q4_0_QUANTIZED
# Run inference directly in the terminal:
./llama-cli -hf kevinharry/SmolLM2-135M-4-bit:Q4_0_QUANTIZED

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf kevinharry/SmolLM2-135M-4-bit:Q4_0_QUANTIZED
# Run inference directly in the terminal:
./build/bin/llama-cli -hf kevinharry/SmolLM2-135M-4-bit:Q4_0_QUANTIZED

Use Docker

docker model run hf.co/kevinharry/SmolLM2-135M-4-bit:Q4_0_QUANTIZED

LM Studio
Jan

Ollama

How to use kevinharry/SmolLM2-135M-4-bit with Ollama:

ollama run hf.co/kevinharry/SmolLM2-135M-4-bit:Q4_0_QUANTIZED

Unsloth Studio

How to use kevinharry/SmolLM2-135M-4-bit with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for kevinharry/SmolLM2-135M-4-bit to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for kevinharry/SmolLM2-135M-4-bit to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for kevinharry/SmolLM2-135M-4-bit to start chatting

Docker Model Runner
How to use kevinharry/SmolLM2-135M-4-bit with Docker Model Runner:
```
docker model run hf.co/kevinharry/SmolLM2-135M-4-bit:Q4_0_QUANTIZED
```

Lemonade

How to use kevinharry/SmolLM2-135M-4-bit with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull kevinharry/SmolLM2-135M-4-bit:Q4_0_QUANTIZED

Run and chat with the model

lemonade run user.SmolLM2-135M-4-bit-Q4_0_QUANTIZED

List all available models

lemonade list

SmolLM2-135M-4-bit

This repository contains a 4-bit quantized version of the HuggingFaceTB/SmolLM2-135M model, using the q4_0 quantization method from llama.cpp, stored in the GGUF file format. Quantization reduces the model's size and memory footprint while maintaining its core capabilities, making it suitable for deployment on resource-constrained environments such as edge devices, mobile platforms, or lightweight inference tasks.

Quantization Details: Base Model: HuggingFaceTB/SmolLM2-135M Quantization Method: q4_0 (4-bit) Framework Used: llama.cpp File Format: GGUF

Downloads last month: -

GGUF

Model size

0.1B params

Architecture

llama

Hardware compatibility

4-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support