# Where LM Studio stores models on Mac, Windows and Linux

> LM Studio stores models in a Hugging Face-shaped tree that is easy to read once you know the convention. Where it is on each platform, what the extra files are, and how to share one model store with other tools.

Canonical: https://bytesweep.app/blog/lm-studio-storage.html
Last updated: 2026-10-16

## Where LM Studio keeps models, and how to read the folder

LM Studio’s model folder is the easy one to audit. Unlike Ollama’s content-addressed store, it mirrors how models are published, so the path tells you who made a file and what it is without opening the app.

- By [Shajeen Ahamed](https://bytesweep.app/about.html#author)
- Published 3 October 2026
- Last updated 16 October 2026
- 3 min read
- Part of [Where AI models are stored](https://bytesweep.app/ai-model-storage.html)
- [How we write](https://bytesweep.app/about.html#how-we-write)

| Platform | Location |
| --- | --- |
| macOS | `~/.lmstudio/models`, older versions `~/.cache/lm-studio/models` |
| Linux | `~/.lmstudio/models` |
| Windows | `C:\Users\<you>\.lmstudio\models` |

The app shows the current path under My Models, and you can change it there. That’s the supported way to move the store. On a Mac that has been upgraded through several versions, check both paths: the old one isn’t cleaned up automatically, and it’s common to find tens of gigabytes in a directory the app no longer reads.

```
du -sh ~/.lmstudio/models ~/.cache/lm-studio 2>/dev/null
```

### Reading the tree

The layout is `publisher/repository/file`, and most of what you need to decide about a file is in its name. The parameter count gives the scale; the suffix gives the quantisation. `Q4_K_M` is a four-bit mixed quantisation, `Q8_0` is eight-bit, and `F16` is unquantised half precision. The same model at `F16` is a bit over three times the size of its `Q4_K_M`, and what quantisation actually costs you covers choosing between them.

*Most of the folder is weights you chose; the waste hides in partial downloads and second copies in another format.*

A few kinds of file sit beside the weights, and they need different treatment.

Split models are named `...-00001-of-00003.gguf`. They’re one model, not three, and every part is needed. Delete one and the rest are useless, so remove the whole repository folder or nothing.

Vision models come with a projector file, usually named `mmproj-...gguf`. It’s small next to the weights, but the model can’t read images without it.

Interrupted downloads can leave partial files behind, often with a suffix such as `.part` or `.incomplete`. Those are pure waste and you can delete them without a second thought:

```
find ~/.lmstudio/models -name '*.part' -o -name '*.incomplete' | xargs -r du -sh
```

MLX models, if you use the MLX runtime on Apple silicon, are folders of `.safetensors` files plus config and tokeniser JSON. An MLX folder next to a GGUF of the same model is a second full copy in a different format, and it’s the commonest accidental duplicate in this folder.

To see what’s taking the space, largest last:

```
find ~/.lmstudio/models -name '*.gguf' -exec du -h {} + | sort -h | tail -20
```

LM Studio reads the directory each time it looks for models; there’s no index to keep in sync. Deleting a folder in the Finder is as good as deleting it in the app.

### Can other tools use these files?

Sometimes. It depends on the format each tool reads.

llama.cpp reads GGUF, so you can point it straight at a file in LM Studio’s tree and skip the second copy.

Ollama keeps its own content-addressed blob store. It can import a GGUF through a Modelfile, but the import copies the file, so you end up with two. [Where Ollama keeps its own store](https://bytesweep.app/blog/ollama-models-disk.html) explains the layout.

Hugging Face transformers wants the safetensors repository, not a GGUF. That’s why machines doing both local inference and Python work end up with the same model twice, once here and once in [the Hugging Face cache](https://bytesweep.app/blog/huggingface-cache-cleanup.html).

When disk is tight, I’d pick one runtime per model family rather than trying to make every tool read one copy. Clear partial downloads first, then any quantisation you’ve superseded: if you have `Q8_0` and `Q4_K_M` of an 8B model and only use the smaller one, the larger is about 8.5 GB doing nothing. Deciding which local models to delete covers the harder calls.

An external drive works for this folder. Over Thunderbolt or USB 3.2 the only cost is a slower load, since the weights are read once and then live in memory.

### Let Bytesweep clear this for you

Skip the commands. Bytesweep finds this for you, rates every item safe, review first or keep, and clears what you choose. Most of what it removes can be put back for seven days. See what it finds on your own disk, free for 3 days, then pay once with no subscription.

[Try it free for 3 days](https://bytesweep.app/download.html) [Buy · $29.99](https://bytesweep.app/pricing.html)

Related in Bytesweep: [Developer tools](https://bytesweep.app/dev-servers.html)

### Keep reading

- [**Where Ollama stores models on Mac, Linux and Windows** Ollama's model store is often the largest folder on a developer's machine, and it lives in a hidden folder the Finder does not show. The exact paths on macOS, Linux, Windows and Docker, how to find the one your install really uses, how to delete models safely and how to move the store to another drive.](https://bytesweep.app/blog/ollama-models-disk.html)
- [**The Hugging Face cache keeps every revision you ever pulled** The hub cache is content-addressed, symlinked and versioned, which makes it efficient and makes it grow. How to read the structure, what hf cache prune and rm do, and why deleting blobs by hand breaks it.](https://bytesweep.app/blog/huggingface-cache-cleanup.html)
- [**Moving AI model stores to an external drive, properly** Every local AI tool keeps its weights somewhere different and each has a supported way to relocate them. The environment variables and settings that work, the file system that will break symlinks, and the performance you actually lose.](https://bytesweep.app/blog/move-ai-models-external-drive.html)
