# meta-llama/llama-guard-4-12b — End-of-Life — shutdown 03/05/26

For meta-llama/llama-guard-4-12b, provider is Groq; lifecycle status is End-of-Life; deprecation announced is February 10, 2026; shutdown / retirement date is 03/05/26; recommended replacement is openai/gpt-oss-safeguard-20b, recorded from its source on 2026-09-09.

- **Provider:** Groq _(our reading, not quoted from the source)_
- **Model ID / API name:** meta-llama/llama-guard-4-12b _(verified: appears in the quote below)_
- **Lifecycle status:** End-of-Life
- **Deprecation announced:** February 10, 2026 _(verified: appears in its own passage below)_
- **Shutdown / retirement date:** 03/05/26 _(verified: appears in the quote below)_
- **Recommended replacement:** openai/gpt-oss-safeguard-20b _(verified: appears in the quote below)_
- **Platform / deployment scope:** Together AI serverless inference _(per docs.together.ai, not stated by the source above)_

## What the source says

> March 5, 2026: meta-llama/llama-guard-4-12b In line with our commitment to bringing you cutting-edge models, on February 10, 2026, we emailed users to announce the deprecation of meta-llama/llama-guard-4-12b in favor of openai/gpt-oss-safeguard-20b. The new GPT-OSS-Safeguard model delivers policy-following reasoning capabilities for Trust & Safety workflows, enabling customizable content moderation with bring-your-own-policy support. Deprecated Model | Shutdown Date | Recommended Replacement Model ID | meta-llama/llama-guard-4-12b | 03/05/26 | openai/gpt-oss-safeguard-20b |

## Where each value comes from

This source states these in separate places, so each value is shown with the passage that states it.

### Deprecation announced

> In line with our commitment to bringing you cutting-edge models, on February 10, 2026, we emailed users to announce the deprecation of meta-llama/llama-guard-4-12b in favor of openai/gpt-oss-safeguard-20b. The new GPT-OSS-Safeguard model delivers policy-following reasoning capabilities for Trust & Safety workflows, enabling customizable content moderation with bring-your-own-policy support.

## Sources disagree

More than one authority states this, and they do not state the same thing. Both are reproduced with the source each came from.

### Provider

console.groq.com says provider is **Groq**, as of 2026-09-09.

> March 5, 2026: meta-llama/llama-guard-4-12b In line with our commitment to bringing you cutting-edge models, on February 10, 2026, we emailed users to announce the deprecation of meta-llama/llama-guard-4-12b in favor of openai/gpt-oss-safeguard-20b. The new GPT-OSS-Safeguard model delivers policy-following reasoning capabilities for Trust & Safety workflows, enabling customizable content moderation with bring-your-own-policy support. Deprecated Model | Shutdown Date | Recommended Replacement Model ID | meta-llama/llama-guard-4-12b | 03/05/26 | openai/gpt-oss-safeguard-20b |

Source: https://console.groq.com/docs/deprecations

docs.together.ai says provider is **Together AI**, as of 2026-09-15.

> | 2026-08-25                  | `meta-llama/Llama-Guard-4-12B`                      | No                                         |

Source: https://docs.together.ai/docs/deprecations.md


### Lifecycle status

console.groq.com says lifecycle status is **End-of-Life**, as of 2026-09-09.

> March 5, 2026: meta-llama/llama-guard-4-12b In line with our commitment to bringing you cutting-edge models, on February 10, 2026, we emailed users to announce the deprecation of meta-llama/llama-guard-4-12b in favor of openai/gpt-oss-safeguard-20b. The new GPT-OSS-Safeguard model delivers policy-following reasoning capabilities for Trust & Safety workflows, enabling customizable content moderation with bring-your-own-policy support. Deprecated Model | Shutdown Date | Recommended Replacement Model ID | meta-llama/llama-guard-4-12b | 03/05/26 | openai/gpt-oss-safeguard-20b |

Source: https://console.groq.com/docs/deprecations

docs.together.ai says lifecycle status is **Removed**, as of 2026-09-15.

> | 2026-08-25                  | `meta-llama/Llama-Guard-4-12B`                      | No                                         |

Source: https://docs.together.ai/docs/deprecations.md


### Shutdown / retirement date

console.groq.com says shutdown / retirement date is **03/05/26**, as of 2026-09-09.

> March 5, 2026: meta-llama/llama-guard-4-12b In line with our commitment to bringing you cutting-edge models, on February 10, 2026, we emailed users to announce the deprecation of meta-llama/llama-guard-4-12b in favor of openai/gpt-oss-safeguard-20b. The new GPT-OSS-Safeguard model delivers policy-following reasoning capabilities for Trust & Safety workflows, enabling customizable content moderation with bring-your-own-policy support. Deprecated Model | Shutdown Date | Recommended Replacement Model ID | meta-llama/llama-guard-4-12b | 03/05/26 | openai/gpt-oss-safeguard-20b |

Source: https://console.groq.com/docs/deprecations

docs.together.ai says shutdown / retirement date is **2026-08-25**, as of 2026-09-15.

> | 2026-08-25                  | `meta-llama/Llama-Guard-4-12B`                      | No                                         |

Source: https://docs.together.ai/docs/deprecations.md


## Source

- https://console.groq.com/docs/deprecations
- https://docs.together.ai/docs/deprecations.md

Last verified: 2026-09-09. Review by: 2026-10-09.
Part of [AI model deprecation and retirement dates by provider](https://referencesource.org/ai-model-deprecation-and-retirement/).