Offline AI File Renamer with Zush

Rename files offline on Mac or Windows with Zush and a local LM Studio or Ollama model. Download and test a vision-capable model first, then preview content-based filenames, apply the batch, and use Rename History to undo changes. Start with 50 free renames shared across all four AI modes.

Download

Get started for free
No registration
No credit card required

IMG_4382.HEIC

Golden Retriever Park.heic

Screenshot 2026-03-23.png

Figma Checkout Flow.png

scan_0042.pdf

Apartment Lease 2026.pdf

meeting_notes_FINAL.docx

Client Discovery Notes.docx

demo_take_02.mov

Settings Sidebar Walkthrough.mov

track_01_final.mp3

Lo-Fi Piano Loop 92BPM.mp3

Set up Zush for offline file renaming

Use the same Templates, preview, and Rename History while a model on your computer analyzes supported files. Prepare the model while online, then test a small batch before taking the workflow offline.

Install Zush and LM Studio or Ollama, then download a compatible vision-capable model. Check its memory and disk requirements; there is no single hardware minimum that fits every model.
Start the model server. In Zush Settings → AI Modes, choose LM Studio or Ollama, select the installed model, and run Test. Use the loopback endpoint so analysis stays on this computer.
Add copies of a few screenshots, photos, or document previews. Review the suggested titles, apply selected names, and test Rename History. Verify the same small workflow without a network connection before relying on it offline.

Zush with LM Studio

Download and load a vision-capable model. In LM Studio, open Developer and start the local server. In Zush AI Modes, select LM Studio, confirm http://127.0.0.1:1234, refresh the model list, select the model, and run Test.

Follow the LM Studio setup

Zush with Ollama

Install Ollama and download a vision-capable model. Start Ollama, then select it in Zush AI Modes. Confirm http://127.0.0.1:11434, refresh the installed models, select your model, and run Test.

Follow the Ollama setup

Zush AI Modes on Mac with LM Studio selected, the local host 127.0.0.1:1234, a selected model, and a successful connection test. — Existing Mac product screenshot: LM Studio is selected and the connection test reports success. The model shown is an example configuration, not a required model or an offline performance benchmark.

scan_0042.pdfNorthstar_Invoice-1042.pdf

Illustrative naming pattern: supplier + invoice number. Check every identifier against the scan. This example is not a measured local-model result.

FREE includes 50 renames shared across all four AI modes; PRO removes the app limit. Model downloads and app updates require internet. Speed and supported analysis depend on the model, file type, and hardware. BYOK uses a cloud provider, and a remote model-server URL is not on-device processing. Zush does not silently switch to cloud analysis while a local mode is selected.

Take a tour of Zush

One quick showcase per feature. Click a tab to see it in action.

Bulk rename files by real content before you apply changes

Your naming system, powered by AI

Combine file content, metadata, dates, custom instructions, and reusable blocks into filenames that match the way you work. Zush suggests — you review and decide.

Smart Batch Rename

Analyze and rename 100+ file formats in one batch, from images, EPUB and FB2 ebooks, and PDFs to audio, video, and Office files. Describe any detail for Zush to extract, then reuse it as a Custom AI Block in any template.

Choose where AI runs

Choose Zush Cloud, BYOK, LM Studio or Ollama for the same review-first rename workflow.

Cloud AI

Goes through our relay to a commercial AI provider, under their standard API terms. The relay exists so the API key can’t be pulled out of the app.

Bring Your Own Key

Connect your own API key from Gemini, OpenRouter, OpenAI, or Claude for cloud renames; FREE includes 50 across all modes and PRO is unlimited. Keys are stored locally in secure platform storage.

Private Local AI Options

Keep supported analysis on your device with LM Studio or Ollama. Each local mode works offline after setup.

Watch one or multiple folders — including iCloud Drive, Google Drive, Dropbox, and OneDrive folders. Zush runs in the background and processes new files automatically.

Supported File Formats

Images

pngjpgjpegjpewebpavifgifbmpdibtifftifheicheicsheifsvgpdfjp2jpfjpxj2kj2cexrtgaicoicnspbmpgmppmpfmcr2cr3nefarwdngorfrafrw2pefsrwsr2raw

Documents

pdftxtmdjsonemlcsvtsvlogxmlyamlymlsrtvttrtfodtodsodppagesnumberskeydocdocxdocmdotxdotmpptpptxpptmpotxpotmppsxppsmxlsxlsxxlsmxltxxltmepubfb2fb2.zip

Videos

mp4movqtm4vmpegmpgm2v3gp3gpp3g23gp2dvtsmtsm2tsvob

Audio

mp3m4awavflacoggwebmmpga

Design

sketchfigaipsd

Try Zush free

Rename and organize files with AI, monitor folders, and clean up screenshots, design files, PDFs, photos, audio, videos, iWork, and Office documents by content.

Download for Mac

Download for Windows

Free · No credit card required · macOS Sequoia+ · Windows 10+

Available on both desktops

Pick your platform — same Zush, native everywhere

A signed and notarized .dmg, Mac App Store build, or Homebrew cask for macOS. A Microsoft Store install for Windows. Same renaming engine, same BYOK and Local AI mode, native to each OS.

Zush for Mac

Apple Silicon and Intel · macOS 15+ · App Store + .dmg + Homebrew

See Mac details

Zush for Windows

x64 and ARM64 · Windows 10/11 · Microsoft Store

See Windows details

Frequently Asked Questions

How can I rename files with AI offline?

Use Zush with LM Studio or Ollama on Mac or Windows. Download a compatible vision model, start its local server, select that mode in Zush AI Modes, and test the connection. Add a small batch, review suggested names, and apply the selected changes with Rename History available.

What do I need to download before going offline?

Install Zush and LM Studio or Ollama, then download a compatible vision-capable model while online. Check the model requirements for memory and disk space. Load the model and test your actual file types before relying on the setup without internet.

How do I connect Zush to LM Studio or Ollama?

Start the model server and open Zush Settings → AI Modes. For LM Studio, the default host is http://127.0.0.1:1234; for Ollama it is http://127.0.0.1:11434. Select an installed vision-capable model and run Test. Keep the endpoint on your own computer for on-device analysis.

Is BYOK an offline AI mode?

No. BYOK sends analysis to your selected cloud provider using your own key. LM Studio and Ollama use the model server you configure; the default loopback endpoints keep supported analysis on your computer. A remote server URL is not on-device processing.

Is offline AI file renaming free in Zush?

Zush FREE includes 50 renames shared across Cloud AI, BYOK, LM Studio, and Ollama. Switching to a local mode does not reset or remove that limit. PRO removes the app limit; local models use your own hardware.

Does offline AI work equally well for every file type?

No. Results and speed depend on the model, available memory, file type, and the content Zush can provide for analysis. Use a vision-capable model for screenshots, scans, and document previews. Verify a small sample of each file type; support in the app does not guarantee identical results across AI modes.

What if the local model is unavailable?

Start the local server, load the downloaded vision model, refresh the model list, and run Test again. Zush does not silently switch to cloud analysis while a local mode is selected. Downloads and updates still require internet; local analysis is not a claim that every app service is permanently offline.

Can I undo offline AI renames?

Yes. Review suggested names before applying a batch, then use Rename History to inspect original and new filenames and revert an applied batch. Test the recovery workflow on copied files before enabling folder monitoring.

Choose your PRO plan

Monthly flexibility or one-time lifetime access. Both plans remove the rename limit across every AI mode.

Trusted by 1,500+ users

PRO Monthly

Flexible monthly plan

$10/month +VAT

Unlimited AI Renames

PRO One-Time

Pay once for lifetime access

~~$48~~$36one-time +VAT

Limited Offer25% offwith code