TripShow TripShow

Documentation & Setup Guide

Everything TripShow can do — and how to turn on AI captions.

Getting started

Choosing & ordering photos

Analyze

Project Settings

The Project Settings button (toolbar) opens a popover of options saved inside this document. It starts from your global defaults and overrides them per project.

Captions

⚠️ With Ollama or OpenRouter, the photo is sent to that model to write the caption. Ollama is your own local server; OpenRouter is a cloud service, so on that path images leave your Mac — TripShow asks you to confirm and shows a red Privacy alert bar while it's active.

Setting up AI captions

Captions need an AI backend. Pick one, configure it in Settings ▸ AI providers, then select it in Project Settings ▸ AI provider. Every option needs a vision-capable model — one that can actually look at the photo. A text-only model will not work.

On-device · free

Apple Intelligence

Enable Apple Intelligence in System Settings ▸ Apple Intelligence & Siri. Nothing to install — captions are generated on your Mac. Availability depends on region and Apple Silicon.

Local · free

Ollama

Install the app from ollama.com, then pull a vision model in Terminal:

ollama pull gemma4:12b

Other vision models: llava, llama3.2-vision (browse ollama.com/library). Keep Ollama running, then in Settings set the Base URL (default http://localhost:11434) and the model name, and click Test Connection. Everything stays on your Mac.

Cloud · paid

OpenRouter

Create an account and an API key at openrouter.ai (keys at openrouter.ai/keys; add credits to your account). Paste the key into Settings and set a vision model such as openai/gpt-4o-mini or google/gemini-flash-1.5 (browse openrouter.ai/models).

⚠️ Privacy — OpenRouter shares your photos. Every photo you caption is uploaded to OpenRouter and on to the model provider you choose; the images leave your Mac. Only use it for photos you're comfortable sharing. For anything private, use Apple or Ollama, which caption locally.

Which model? A bigger local model like gemma4:12b gives richer captions but needs more RAM and runs slower than a small one like llava. Cloud models are fast and capable but cost per image and send your photos to the provider.

↑ Back to top

Managing sources

A project can pull from several folders and Photos albums at once. More ▸ Manage Sources… is where you review and prune them.

Music & analysis

Play

Export

Settings vs. Project Settings

Privacy