Your Machine. Your Models. Your Control.

⚡

How it works

Your tools, your data, your way

Local first

NodeTool runs your workflows on your machines without relying on any vendor cloud infrastructure.

Open source

NodeTool is open source, so you can inspect, modify, and self-host the entire stack.

Data stays yours

NodeTool processes data locally and never collects, phones home, or sends usage telemetry anywhere.

No lock in

NodeTool keeps one portable workflow format from laptop to deployment, so you control every environment.

Use cases

Build workflows for any task

Smart assistants

Build chatbots that search documents, browse websites, and answer questions.

Agent workflows

Chain multiple AI models to research topics, write reports, and organize findings.

Automation

Schedule tasks, process files in batches, and trigger workflows based on events.

Content generation

Generate images, videos, and written content with text prompts and style controls.

Voice and audio

Transcribe recordings, synthesize speech, and remove background noise from audio files.

Data analysis

Classify text, detect objects in images, and extract insights from large datasets.

Explore templates →

Deploy when you need scale

Same workflow file runs locally or on serverless GPUs with no rewrites.

Run one command to deploy. NodeTool provisions infrastructure, downloads models, and configures containers automatically. Endpoints scale to zero when idle. Works with RunPod and Google Cloud. Self-host on your own infrastructure.

RunPod Google Cloud

Self-Hosted

GPU usage is billed directly by the provider.

⚡

Fast local inference

Native support for MLX (Apple Silicon), GGUF, and more.

🍎

Apple SiliconM1-M4

MLX

Apple Silicon optimized: LLMs, audio, speech, and image generation with mflux

MLX-LM
MLX-Audio
MLX-Whisper
mflux

Learn more

🦙

Cross-Platform87k+ stars

llama.cpp + whisper.cpp

Fast LLM inference and speech recognition on any platform

GGUF models
CPU & GPU
Speech-to-text

Learn more

📦

Inference EngineProduction-grade

vLLM

Production-grade high-throughput inference engine

High throughput
Paged attention
Continuous batching

Learn more

🤗

HuggingFace — Transformers & Diffusers

Run thousands of HuggingFace models locally with our native integration.

Text Generation

Image-to-Image

Speech Recognition

Text-to-Speech

Image Classification

Object Detection

Summarization

Translation

Question Answering

Text-to-Image

Video Processing

Depth Estimation

Image Segmentation

Audio Classification

Token Classification

Feature Extraction

Sentence Similarity

Multimodal

📦Explore nodetool-huggingface Package

24+ model types • Thousands of models • All running locally

🧠

Models

Model Manager

Download and manage model weights from Hugging Face.

🔗

Use any provider you want

Bring your own API keys for OpenAI, Anthropic, Gemini, Fal AI, Replicate, and HuggingFace Inference Providers. No markup, no middleman. Or skip APIs entirely and use local models.