Technický PRŮVODCE

Llama Ochranné a bezpečnostní klasifikátory

Llama Guard is a family of safety classifiers that labels prompts and model responses against a defined hazard policy.

Poslechněte si tento návodAI-vyprávěný · ~3 min

Na této stránce3 min čtení
  1. Přehled
  2. Hluboký ponor
  3. Real-World Implementace
  4. The Future of Llama Guard and Safety Classifiers
  5. Uvedení do praxe
  6. Pokračujte v objevování
  7. Často kladené otázky

Přehled

The current Llama Guard 4 12B release is a multimodal classifier for text and images, while earlier versions have different capabilities; a classifier result is one safety signal, not a complete moderation or security system.

Hluboký ponor

Llama Guard is a safety-classification model intended to work alongside a generative model. It can classify an incoming prompt, a generated response, or both, and returns a safe/unsafe label with category information for unsafe content. That output can support a product’s moderation decision, but the application still has to define what to do with each label: block, transform, escalate, or permit. A classifier’s categories and training policy are not automatically identical to an organization’s legal duties or product rules. Capabilities depend on the version. Meta’s current Llama Guard 4 12B model card describes a natively multimodal classifier for text and one or more images, based on a hazard taxonomy, with prompt and response classification. It also describes an added text-only code-interpreter-abuse category. Earlier Guard versions are different; do not transfer Guard 4’s image support, parameter size, or language claims to them. The official documentation gives model-specific prompt formats and compatibility guidance, so deployments should verify the exact checkpoint and interface they use. Use input filtering when a product wants to stop certain requests before generation, and output filtering when it wants to inspect what the model actually produced. Using both provides two decision points, but can also increase refusals or latency. Meta documents limitations including imperfect policy coverage, dependence on training knowledge for some categories, and susceptibility to adversarial or prompt-injection attacks. Evaluate the full application with representative examples, inspect false positives and misses, and add other controls for sensitive decisions. A safety label is evidence for a workflow, not proof that all unsafe content has been caught.

Real-World Implementace

An application checks a user prompt before generation, then separately checks the generated response before showing it, recording which stage triggered a block.

A team handling mixed text-and-image requests evaluates both parts together with Llama Guard 4 and tests its behavior on the exact image formats used by the application.

A product maps its moderation policy to the classifier’s supported hazard categories and sends uncovered cases to a separate review path.

A multilingual deployment tests representative languages, policy edge cases, false positives, and adversarial inputs before deciding where the model is suitable.

The Future of Llama Guard and Safety Classifiers

Safety classifiers are likely to remain useful as modular checks in systems that combine models, policy code, and human review. Multimodal checkpoints widen the kinds of input a single classifier can inspect, but broader coverage does not remove evaluation gaps or deployment-specific risks. Teams should track model-card changes, repeat tests when versions or policies change, and avoid treating vendor benchmark results as a substitute for testing their own use case. Publish internal acceptance criteria for the application’s actual language and modality mix.

Uvedení do praxe

Proč na tom záleží

  • Cena a rozpočet. Rozhodnutí o architektuře zvyšují výkon a provozní náklady po mnoho let.
  • Jasnější rozhodnutí. Technické vzdělání pomáhá týmům vybrat ten správný stack, nejen ten nejnovější.
  • Kontrola kvality. Lepší konstrukční volby snižují výskyt problémů se spolehlivostí ve výrobě.

Pozor na

  • Optimalizace jednoho benchmarku může skrýt širší systémové slabiny.
  • Náklady na infrastrukturu a údržbu jsou často podceňovány.
  • Mezery v zabezpečení a pozorovatelnosti se mohou zvětšovat, jak se systémy stávají složitějšími.

Další kroky

  1. Před implementací definujte cíle latence, kvality a nákladů.
  2. Benchmark za realistických podmínek zatížení a dat.
  3. Monitorování chyb, posunu a dopadu na uživatele.
  4. Před škálováním připravte cesty vrácení zpět a reakce na incidenty.

Pokračujte v objevování

Newsletter zdarma

Získejte denní přehled o AI

Tři ověřené příběhy ze světa AI každý všední den ráno, psané srozumitelnou angličtinou. Navždy zdarma, bez reklam.

Jeden e-mail každý všední den. Odhlásit se můžete jedním kliknutím. Vaši adresu nikdy neprodáváme ani nesdílíme.

Otestujte se

Take the Llama Guard and Safety Classifiers quiz

Okamžitá zpětná vazba ke každé odpovědi a sdílitelný certifikát s ověřitelným ID po absolvování kurzu.

Spustit kvíz

Podpořte bezplatnou AI vzdělávání. AI Understanding je nezisková organizace 501(c)(3) — žádné reklamy, žádná platba za přístup, nikdy. Darujte

Často kladené otázky

What is Llama Guard and Safety Classifiers?

Llama Guard is a family of safety classifiers that labels prompts and model responses against a defined hazard policy. The current Llama Guard 4 12B release is a multimodal classifier for text and images, while earlier versions have different capabilities; a classifier result is one safety signal, not a complete moderation or security system.

What does a Llama Guard classifier provide to an application?

The model card describes labels for safe or unsafe content and hazard categories; application policy still determines the next action.

Which modalities does Meta describe for Llama Guard 4 12B?

Meta describes Guard 4 as natively multimodal and classifying text together with one or more images; it is not image-only classification.

How should a team use Llama Guard 4 image support when reviewing a mixed request?

The current card describes joint text-and-image input and says the model is not designed for image-only classification.

How does input filtering differ from output filtering?

The model card distinguishes prompt classification before generation from response classification after generation.

Why might a system combine input and output checks?

Meta describes both filtering modes and notes that using both gives additional security, while limitations remain.