Home / Photography / Llama for photography
Meta
PhotographyLlama for photography
Meta's natively multimodal, open-weight Mixture-of-Experts family (Scout and Maverick), free to download and self-host.
Llama is not on our list for photography. Usable for general work, but first look at the options we recommend for photography.
- GDPR profileOur summary of what the provider itself publishes about your data: EU hosting possible, no training on your data for business use, and self-hosting available. Caution applies to a model with a documented problem, such as an investigation by a regulator. How we measure
- GDPR: Caution
- QualityQuality from blind duels on LMArena: users choose between two anonymous answers. A difference of about 20 points is noticeable, a few points is noise. How we measure
- –Elo, independent
- Editorial scoreOur own assessment on 10 of suitability for professional work, price and privacy. This is a judgement, not a measurement; that is why it always sits alongside the Elo. How we measure
- 7,6
- Entry price
- Free
What Llama does in photography
- Culling photos: sharpness, eyes, duplicates and the best shot per moment
- Editing in your own style: applying AI profiles and presets across a whole shoot
- Retouching and enhancing: skin, background, noise, sharpness and cropping in batches
- Archiving and delivering: keywords, face search and client galleries
- Open weights, fully self-hostable
- Fine-tunable on your own data
- Very large context window (Scout 10M)
- GPU-efficient (Mixture-of-Experts)
Llama has a specific GDPR concern. The Llama 4 Acceptable Use Policy prohibits the multimodal models for businesses headquartered in the EU. Dutch businesses may therefore not install the multimodal Llama 4 models themselves under the licence. Weigh this heavily before using it in photography.
Confidentiality and GDPR
GDPR: CautionYour photos show recognisable people, often clients and their guests: that is personal data, and as soon as a tool groups or recognises faces, it becomes biometric data (Art. 9 GDPR). So check whether a tool works locally or uploads your photos. Aftershoot, Narrative and Excire select on your own computer; Excire works without the cloud. Neurapix processes in Göttingen on its own hardware and deletes images up to 48 hours after delivery. Watch out for training: Aftershoot states in its privacy statement that collected images and edits are used to train its models, Pixelz uses images for machine learning with the client's consent, and Pic-Time trains on a small share of images from photographers who sign up for this. Luminar Neo sends photos for advanced AI features to external AI partners, including outside the EEA. If you use facial recognition in client galleries, record that in your terms and sign a data processing agreement.
Self-hosting gives full data control within your own EU infrastructure, a strong GDPR argument.
Prices
Other models for photography
Claude for photographyAnthropic · editorial 9.01506
Gemini for photographyGoogle · editorial 8.61527
ChatGPT for photographyOpenAI · edition 8.91485
Le Chat (Mistral) for photographyMistral AI · editorial 8.41426We don't know of a professional tool for photography that runs publicly on Llama.
Questions about Llama for photography
- Is Llama suitable for photography?
- Not specifically. Llama is not on our list for photography; it is broadly applicable, but first look at the options we recommend for photography. Keep a professional in the loop for every use.
- May I put confidential data into Llama?
- Your photos show recognisable people, often clients and their guests: that is personal data, and as soon as a tool groups or recognises faces, it becomes biometric data (art. 9 GDPR). So check whether a tool works locally or uploads your photos. Aftershoot, Narrative and Excire select on your own computer; Excire works without the cloud. Neurapix processes in Göttingen on its own hardware and deletes images no later than 48 hours after delivery. Mind the training use: Aftershoot states in its privacy statement that collected images and edits are used to train its models, Pixelz uses images for machine learning with the client's consent, and Pic-Time trains on a small share of images from photographers who opt in. Luminar Neo sends photos for advanced AI features to external AI partners, including outside the EEA. If you use facial recognition in client galleries, record that in your terms and sign a data processing agreement. Self-hosting gives full data control within your own EU infrastructure, a strong GDPR argument.
- What does Llama cost?
- There is self-hosting (free), Meta AI ($0), Hosted API (third parties) (from ~$0.03 / 1M in). Check current prices on the official site.
Every Friday: what changed in AI for photography
New models, prices, oversight and tools, with sources and what it means for you. One email a week, unsubscribe with one click.




