Instructions to use the-robot-ai/TinyLink with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use the-robot-ai/TinyLink with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="the-robot-ai/TinyLink",
	filename="TinyLink.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 the-robot-ai/TinyLink with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf the-robot-ai/TinyLink
# Run inference directly in the terminal:
llama-cli -hf the-robot-ai/TinyLink

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf the-robot-ai/TinyLink
# Run inference directly in the terminal:
llama-cli -hf the-robot-ai/TinyLink

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 the-robot-ai/TinyLink
# Run inference directly in the terminal:
./llama-cli -hf the-robot-ai/TinyLink

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 the-robot-ai/TinyLink
# Run inference directly in the terminal:
./build/bin/llama-cli -hf the-robot-ai/TinyLink

Use Docker

docker model run hf.co/the-robot-ai/TinyLink

LM Studio
Jan
Ollama
How to use the-robot-ai/TinyLink with Ollama:
```
ollama run hf.co/the-robot-ai/TinyLink
```

Unsloth Studio new

How to use the-robot-ai/TinyLink 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 the-robot-ai/TinyLink 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 the-robot-ai/TinyLink to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for the-robot-ai/TinyLink to start chatting

Docker Model Runner
How to use the-robot-ai/TinyLink with Docker Model Runner:
```
docker model run hf.co/the-robot-ai/TinyLink
```

Lemonade

How to use the-robot-ai/TinyLink with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull the-robot-ai/TinyLink

Run and chat with the model

lemonade run user.TinyLink-{{QUANT_TAG}}

List all available models

lemonade list

TinyLink

📌 Summary

TinyLink is a lightweight language model fine-tuned to translate natural language instructions into commands for controlling drones and robots via MAVLink. It is designed for edge robotics. Unlike solutions relying on cloud APIs, TinyLink runs fully offline on your local machine.

Demo & Instructions

For a demo on how to use this model, you can check the following Github repo

Features

Translates plain text instructions into MAVLink commands.
Runs entirely on-device for enhanced privacy. No API keys or cloud dependency.
Runs on everyday hardware; no GPU or excessive RAM needed.
Tested with ArduPilot SITL.
Achieves 0.9–2.2s inference times on CPU, depending on hardware.
Supported Commands:
1. Arm
2. Disarm
3. Takeoff
4. Land
5. Change mode (limited modes supported)
6. Move in X, Y, Z (Copter and Rover)

Performance & Tested Platforms

Platform	RAM	Inference Time (avg)	Status
Win 11 (App) & WSL2 (SITL)	16 GB	1.7 - 4s (Avg 2.2s)	✅ Tested
Win 11 (TinyLink)	16 GB	0.5 - 1.2s (Avg 0.9s)	✅ Tested
Raspberry Pi 5	4 GB	0.8 - 2s (Avg 1.5s)	✅ Tested
NVIDIA Jetson Nano	-	-	❌ Not tested

Downloads last month: 28

GGUF

Model size

1B params

Architecture

llama

Hardware compatibility

We're not able to determine the quantization variants.

View all variants

Video Preview

Robotics