Meta: Llama Guard 4 12B
Safety model for policy screening, moderation, and risk-aware routing workflows
Context
163,840
Max output
16,384
Input / 1M
$0.18
Output / 1M
$0.18
Providers
3
01Profile
Profile
Released
2025-04-30
Last updated
2026-04-30
Knowledge cutoff
—
Weights
Closed
02Capabilities
Capabilities
Input
image · text
Output
text
Reasoning
No
Tool calling
No
Structured output
No
Attachments
Yes
Cache read
—
Cache write
—
03Pricing
Pricing across 2 providers
Per 1M tokens, USD. Sorted by input price. First-party row highlighted.
04History
Change history
12+ recent events since 2025-04-30
2026-08-25ContextopenrouterContext window1,048,576→ 163,840
2026-08-02ContextkiloMax output32,768→ 16,384
2026-07-21ContextopenrouterContext window163,840→ 1,048,576
2026-06-10Removedgroqofferingremoved→ new
2026-05-15Addedopenrouteroffering→ new
2026-05-07Removeddigitaloceanofferingremoved→ new
2026-05-03Addednvidiaoffering→ new
2026-04-30Addeddigitaloceanoffering→ new
2026-04-18Removedllmgatewayofferingremoved→ new
2026-03-26DeprecatedgroqStatus—→ deprecated
2026-02-16Addedkilooffering→ new
2026-01-23ContextllmgatewayMax output—→ 16,384