This major update introduces the long-awaited Batch Edit dialog, which allows you to edit the metadata of multiple pictures in one go. On the AI side, an upgraded face recognition pipeline delivers more and better matches. PhotoPrism now integrates directly with Ollama and OpenAI to generate captions and labels. Support for custom TensorFlow models, refined configuration, and new scheduling options offer further flexibility.

You'll also notice many performance and usability enhancements, such as the ability to change the cover image for a person, as well as updated dependencies and new translations.

As always, a big thank you to everyone who contributed and helped with testing! We hope you enjoy the new release as much as we do. 🌈💎✨

Upgrade Notes

• To benefit from the facial recognition improvements, we recommend running photoprism faces audit --fix and photoprism faces index in a terminal to resolve any inconsistencies before detecting and matching additional faces. If you want the new engine to re-detect all faces for a clean state, you can do so by executing the commands photoprism faces reset -f and then photoprism faces index (after that, all detected faces must be reassigned). A complete rescan will also detect additional faces, but takes longer since more indexing tasks are performed.
• PhotoPrism now supports multiple AI engines, so the PHOTOPRISM_DISABLE_TENSORFLOW option has been deprecated. You can still disable individual AI features using the feature flags and/or a configuration.

What's new?

• AI: Support for custom TensorFlow image classification models by @raystlin
• AI: Direct Ollama integration for generating captions and labels by @lastzero
• AI: Generate Captions & Labels using the OpenAI Responses API by @lastzero
• AI: Improved face detection and embedding pipeline with a new CNN engine
• AI: Improved model configuration and photoprism vision run command
• AI: Added scheduling options for running vision models in the background

Screenshots

Batch Edit

Quickly and easily update the metadata, albums, and labels of multiple pictures:

Entries that are assigned to all selected pictures are listed first. Each entry is shown as a small chip. If a label or album is assigned to only some of the selected pictures, its chip is shown as partially assigned.

Usability Enhancements

Browse related pictures or set a cover image for a person:

Add pictures to multiple albums in one step:

Ollama AI Model Integration

Use advanced LLMs like the recently released Qwen3-VL to generate captions and labels for your pictures:

Translations

Missing user interface translations have been generated with the help of DeepL and Google Translate. Native speakers are welcome to help us improve them where needed. Learn more ›

Installation Packages

The packages attached to this release are intended for experienced users and maintainers of third-party integrations only, as they require manual configuration and do not include tested system dependencies. Since we are unable to provide support for custom installations, we recommend using one of our Docker images to run PhotoPrism on a private server or NAS device. Learn more ›