Instructions to use llmware/slim-extract-tiny-tool with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use llmware/slim-extract-tiny-tool with Transformers:

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("llmware/slim-extract-tiny-tool", dtype="auto")

llama-cpp-python

How to use llmware/slim-extract-tiny-tool with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="llmware/slim-extract-tiny-tool",
	filename="tiny-extract.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use llmware/slim-extract-tiny-tool with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf llmware/slim-extract-tiny-tool
# Run inference directly in the terminal:
llama-cli -hf llmware/slim-extract-tiny-tool

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf llmware/slim-extract-tiny-tool
# Run inference directly in the terminal:
llama-cli -hf llmware/slim-extract-tiny-tool

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 llmware/slim-extract-tiny-tool
# Run inference directly in the terminal:
./llama-cli -hf llmware/slim-extract-tiny-tool

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 llmware/slim-extract-tiny-tool
# Run inference directly in the terminal:
./build/bin/llama-cli -hf llmware/slim-extract-tiny-tool

Use Docker

docker model run hf.co/llmware/slim-extract-tiny-tool

LM Studio
Jan
Ollama
How to use llmware/slim-extract-tiny-tool with Ollama:
```
ollama run hf.co/llmware/slim-extract-tiny-tool
```

Unsloth Studio new

How to use llmware/slim-extract-tiny-tool 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 llmware/slim-extract-tiny-tool 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 llmware/slim-extract-tiny-tool to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for llmware/slim-extract-tiny-tool to start chatting

Docker Model Runner
How to use llmware/slim-extract-tiny-tool with Docker Model Runner:
```
docker model run hf.co/llmware/slim-extract-tiny-tool
```

Lemonade

How to use llmware/slim-extract-tiny-tool with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull llmware/slim-extract-tiny-tool

Run and chat with the model

lemonade run user.slim-extract-tiny-tool-{{QUANT_TAG}}

List all available models

lemonade list

SLIM-EXTRACT-TINY-TOOL

slim-extract-tiny-tool is a 4_K_M quantized GGUF version of slim-extract, providing a small, fast inference implementation, optimized for multi-model concurrent deployment.

This model has been fine-tuned to implement a general-purpose extraction function that takes a custom key as input parameter, and generates a python dictionary consisting of that custom key with the value consisting of a list of the values associated with that key in the text.

The size of the self-contained GGUF model binary is less than 700 MB, which is small enough to run locally on a CPU with reasonable inference speed, and has been designed to balance solid quality with fast loading and inference on a local machine.

The intent of SLIMs is to forge a middle-ground between traditional encoder-based classifiers and open-ended API-based LLMs.

slim-extract-tiny is part of the SLIM ("Structured Language Instruction Model") series, providing a set of small, specialized decoder-based LLMs, fine-tuned for function-calling.

To pull the model via API:

from huggingface_hub import snapshot_download           
snapshot_download("llmware/slim-extract-tiny-tool", local_dir="/path/on/your/machine/", local_dir_use_symlinks=False)

Load in your favorite GGUF inference engine, or try with llmware as follows:

from llmware.models import ModelCatalog  

# to load the model and make a basic inference
model = ModelCatalog().load_model("slim-extract-tiny-tool")
response = model.function_call(text_sample)  

# this one line will download the model and run a series of tests
ModelCatalog().tool_test_run("slim-extract-tiny-tool", verbose=True)

Note: please review config.json in the repository for prompt wrapping information, details on the model, and full test set.