Inferse
For vendors
AdvertiseTop spots, featured rows, home sponsor Get listedAdd a product we don't cover yet Claim a listingFree: verify the domain, then correct your page Vendor dashboardNumbers, orders and placements
List your platform
Vendor dashboard

Llama Guard 4 12B

Meta· released 30 Apr 2025

meta-llama/llama-guard-4-12b
Input / 1M tokens
$0.18
Output / 1M tokens
$0.18
Context window164K tokens

Longest answer: 16K tokens

reads imagereads textopen weightsvision
See it on openrouter.ai

About

Llama Guard 4 12B is Meta’s multimodal model for content safety classification, derived from Llama 4 Scout and fine-tuned for that purpose. Its listed summary describes classification of content in LLM contexts. It accepts image and text inputs, returns text, and includes vision support. Open weights are available. The model has a context window of 163,840 tokens and a maximum output of 16,384 tokens. Input costs $0.18 per 1M tokens, and output costs $0.18 per 1M tokens. Its model ID is meta-llama/llama-guard-4-12b, and its Hugging Face ID is meta-llama/Llama-Guard-4-12B. Meta released it on 2025-04-30. Teams considering it for a safety-classification role can weigh its image and text inputs, open weights, output bound, and equal listed input and output rates.

Who it is for

It suits teams seeking an open-weight model for content safety classification in LLM contexts. Its image and text inputs may fit classification workflows involving either modality.

What is good

  • Fine-tuned for content safety classification.
  • Open weights are available.
  • Accepts image and text.
  • Input and output each cost $0.18 per 1M tokens.

What to know first

  • Maximum output is 16,384 tokens.
  • Context window is 163,840 tokens.

Verdict

Llama Guard 4 12B is focused on content safety classification and supports both image and text input. Its open weights and equal per-token rates are relevant alongside its 16,384-token output limit.

Details

Lab
Metaopenrouter.ai · 3 Oct 2026
Context
164Kopenrouter.ai · 3 Oct 2026
Input price
$0.18 / 1Mopenrouter.ai · 3 Oct 2026
Output price
$0.18 / 1Mopenrouter.ai · 3 Oct 2026
Max output
16,384 tokensopenrouter.ai · 3 Oct 2026
Inputs
image, textopenrouter.ai · 3 Oct 2026
Open weights
Yesopenrouter.ai · 3 Oct 2026

More longest-context models

See the list
Auto Router ctx2,000KinVaries by model routed tooutVaries by model routed to openrouterAuto Router (Beta) ctx2,000KinVaries by model routed tooutVaries by model routed to openrouterGrok 4.20 ctx2,000Kin$1.25 / 1Mout$2.5 / 1M SpaceXAIGrok 4.20 Multi-Agent ctx2,000Kin$1.25 / 1Mout$2.5 / 1M SpaceXAIPareto Code Router ctx2,000KinVaries by model routed tooutVaries by model routed to openrouterLlama 4 Scout ctx1,311Kin$0.1 / 1Mout$0.3 / 1M Meta

Listed on Inferse

Sources