Pricing · Verified 2026-08-10

Is LM Studio free?

Yes for running models on your own Mac, and that has not changed. What is new is a metered cloud tier beside the free local one, billed per million tokens, so "free" is now the answer about the local app rather than about the whole product.

What costs nothing in LM Studio

  • The desktop app, free for home and work use.
  • Running local models through its llama.cpp and MLX engines.
  • LM Link for up to five devices.
  • Web search with zero data retention.

What LM Studio charges for

Prices as LM Studio publishes them, checked on 2026-08-10. Vendors change pricing without notice, so treat this as a dated reading and check their page before you buy.

TierPriceWhat it buys
Pay as you go$0.13 to $3.00 per million input tokensCloud inference credits. Output is billed separately and runs from $0.26 to $15.00 per million tokens depending on the model.
Bionic PassNot publishedTheir own pages say pricing and plan details are coming soon, so there is nothing to compare yet.

What costs nothing here

The whole native runner is the free tier. No account, no caps, no terminal. This list is read straight from the pricing catalog this site bills from, so it cannot drift from what you actually get.

  • Native llama.cpp engine, runs any GGUF model
  • Unlimited model downloads, multi-model switching
  • Local OpenAI-compatible API + embeddings
  • MCP host: install any MCP server and run its tools through ToolPiper
  • All speech: transcription, text-to-speech, voice cloning, dictation, voice chat
  • Talk to your Mac: spoken commands across 26 system domains
  • Local chat with tools, on any model you run
  • Apple Intelligence on the Neural Engine
  • Browser basics: connect to Chrome, snapshot a page, list tabs
  • Search Console reports: what your site already ranks for
  • Free companion apps (VisionPiper, AudioPiper, MediaPiper)

Where our own paid line sits

The first thing we charge for is Pro, at $10/mo. Nothing above is needed to run models, and nothing above gates anything in the list you just read.

  • Browser automation: act, fill, extract, record, intercept, assert and run Autopilot loops in Chrome
  • Files, Git and archives: the agent reads, edits, searches and commits
  • Clipboard cache tools: the agent reads and writes your clipboard history
  • Mac automation: the agent drives apps, types and clicks, arranges windows and Spaces, launches and stops processes, writes defaults and undoes
  • Local RAG over your files
  • Web scraping and YouTube transcripts
  • Screen and image analysis: screenshot, OCR, color pick, webcam, live stream and capture region controls
  • Image composition: import, fill, outpaint, free transform, perspective, gradients and text
  • Agent audio tools: record, transcribe and speak from a tool call
  • Cloud API proxy (bring your own keys)
  • Outreach toolkit: SEO tools (SERP capture with AI Overview detection, keyword discovery, rank check), Reddit and HN listening, content queue
  • API discovery toolkit + request replay
  • Save captured video streams to disk

Where LM Studio is the better answer

  • LM Studio's model browser is the best in the category and ours is not close.
  • Two inference engines, llama.cpp and MLX, against our one.
  • It runs on macOS 14 or newer, plus Windows and Linux. We need macOS 26 and Apple Silicon, so on an older Mac they work and we do not.
  • Zero data retention on their cloud is a real commitment, not marketing.

Run it and check the list yourself

No account, no caps, no card. Everything in the free list above works on first launch.

Download ToolPiperSee every tier