Instructions to use tbrrss/SmolLM2-135M-Distilled-Q4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tbrrss/SmolLM2-135M-Distilled-Q4 with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="tbrrss/SmolLM2-135M-Distilled-Q4",
	filename="smollm2-135m-distilled-q4.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use tbrrss/SmolLM2-135M-Distilled-Q4 with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf tbrrss/SmolLM2-135M-Distilled-Q4
# Run inference directly in the terminal:
llama-cli -hf tbrrss/SmolLM2-135M-Distilled-Q4

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf tbrrss/SmolLM2-135M-Distilled-Q4
# Run inference directly in the terminal:
llama-cli -hf tbrrss/SmolLM2-135M-Distilled-Q4

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 tbrrss/SmolLM2-135M-Distilled-Q4
# Run inference directly in the terminal:
./llama-cli -hf tbrrss/SmolLM2-135M-Distilled-Q4

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 tbrrss/SmolLM2-135M-Distilled-Q4
# Run inference directly in the terminal:
./build/bin/llama-cli -hf tbrrss/SmolLM2-135M-Distilled-Q4

Use Docker

docker model run hf.co/tbrrss/SmolLM2-135M-Distilled-Q4

LM Studio
Jan

vLLM

How to use tbrrss/SmolLM2-135M-Distilled-Q4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tbrrss/SmolLM2-135M-Distilled-Q4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tbrrss/SmolLM2-135M-Distilled-Q4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/tbrrss/SmolLM2-135M-Distilled-Q4

Ollama
How to use tbrrss/SmolLM2-135M-Distilled-Q4 with Ollama:
```
ollama run hf.co/tbrrss/SmolLM2-135M-Distilled-Q4
```

Unsloth Studio new

How to use tbrrss/SmolLM2-135M-Distilled-Q4 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 tbrrss/SmolLM2-135M-Distilled-Q4 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 tbrrss/SmolLM2-135M-Distilled-Q4 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for tbrrss/SmolLM2-135M-Distilled-Q4 to start chatting

Docker Model Runner
How to use tbrrss/SmolLM2-135M-Distilled-Q4 with Docker Model Runner:
```
docker model run hf.co/tbrrss/SmolLM2-135M-Distilled-Q4
```

Lemonade

How to use tbrrss/SmolLM2-135M-Distilled-Q4 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull tbrrss/SmolLM2-135M-Distilled-Q4

Run and chat with the model

lemonade run user.SmolLM2-135M-Distilled-Q4-{{QUANT_TAG}}

List all available models

lemonade list

SmolLM2-135M-Distilled-Q4

An experimental ~85MB quantized language model distilled from SmolLM2-1.7B-Instruct into SmolLM2-135M.

Try a live demo in your browser, uses wllama.

Details

Teacher: SmolLM2-1.7B-Instruct
Student: SmolLM2-135M-Instruct
Training data: ~50K responses from smoltalk (smol-magpie-ultra, apigen-80k, smol-constraints, openhermes-100k, and smol-summarize)
Method: Response-level knowledge distillation + GGUF Q2_K quantization (fell back to ~Q4_0 because of 576-dim alignment)
Effective quantization: ~5.14 bits/ weight

Usage

llama-cli -m smollm2-135m-distilled-q4.gguf -p "The weather in London is usually" -n 256 --temp 0

Limitations

Well below the instruction-following threshold — generates somewhat coherent text
Does not reliably follow output format constraints
Basic math accuracy is unreliable
Possibly useable for text completion and topic detection, but not structured tasks

Notes

This represents my first attempt at distillation and quantization using RunPod (A40). It is intentionally experimental. Official BitNet materials cautions that models below 3B parameters will be less accurate than full-precision ones. This was a quick proof-of-concept to see how small you could go. Future experiments might yield better results with a higher parameter count and training ternary models from scratch, rather than post-training quantized models.

License

Apache 2.0 (same as SmolLM2)

Downloads last month: 11

GGUF

Model size

0.1B params

Architecture

llama

Hardware compatibility

We're not able to determine the quantization variants.

View all variants

Model tree for tbrrss/SmolLM2-135M-Distilled-Q4

Base model

HuggingFaceTB/SmolLM2-135M

Quantized

HuggingFaceTB/SmolLM2-135M-Instruct

Quantized

(101)

this model