---
title: "What Free includes and what Pro adds"
description: "What is the difference between the free tier and Pro, and do I need to pay?"
date: 2026-09-01
author: "ModelPiper"
tags: ["Pricing", "Tiers", "Pro", "Free"]
type: "faq"
canonical: "https://modelpiper.com/help/pro-versus-free"
---

# What Free includes and what Pro adds

> What is the difference between the free tier and Pro, and do I need to pay?

Free gets you local inference and 327 of the over 420 tools, with no account and no usage caps. Pro is $10 a month and adds the things that reach past your own machine: retrieval over your files, web and search data, and cloud providers through your own API keys.

## What Free includes

-   Any GGUF model on the native engine, with unlimited downloads and switching.
-   A local OpenAI-compatible API and embeddings endpoint.
-   All speech: transcription, text to speech, voice cloning and dictation.
-   Apple Intelligence on the Neural Engine.
-   Browser automation, vision and system control.
-   Search Console reports: what your site already ranks for.
-   The companion apps: VisionPiper, AudioPiper and MediaPiper.

There is no account, no trial clock and no cap on how much you run. Local inference costs us nothing per request, so there is nothing to meter.

## What Pro adds

-   Local retrieval over your own files.
-   Web scraping and YouTube transcripts.
-   A cloud API proxy, using keys you bring.
-   The Outreach toolkit: live search result capture, keyword discovery, rank checking, Reddit and Hacker News listening, and a content queue.
-   API discovery and request replay.
-   Saving captured video streams to disk.

## Do you need it

Not to run models. If what you want is local inference, speech and the tool catalog, Free is the whole product and it stays that way.

Pro is for work that needs the outside world: reading the live market, pulling in your own documents, or routing some requests to a cloud provider you already pay for. Those either cost us something per call or need infrastructure we run, which is the line the tiers are drawn on.

## Speech and Apple's own models are free

Transcription, text to speech, voice cloning and dictation are not metered and not gated, and neither are the on-device Apple Vision and language features ToolPiper exposes. Running on the Neural Engine is not what decides a tier: some Studio features run there too.