# Meta Llama Prompt Guard 2 22M

> Meta Llama Prompt Guard 2 22M is an open-weight language model from Meta, released on 29 Apr 2025. It accepts text and generates text, with a 512-token context window and up to 2 output tokens per response. As of 22 Oct 2026, the lowest listed API price is $0.03 per million input tokens and $0.03 per million output tokens (Groq), across 1 provider we track.

Source page: https://www.aimodel.directory/models/llama-prompt-guard-2-22m
Last verified: 22 Oct 2026

## Specifications

- **Developer:** Meta
- **Release date:** 29 Apr 2025 (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/helicone/models/llama-prompt-guard-2-22m.toml)
- **Status:** Generally available (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/helicone/models/llama-prompt-guard-2-22m.toml)
- **Weights:** Open weights (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/helicone/models/llama-prompt-guard-2-22m.toml)
- **Context window:** 512 (512 tokens) (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/helicone/models/llama-prompt-guard-2-22m.toml)
- **Max output:** 2 tokens (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/helicone/models/llama-prompt-guard-2-22m.toml)
- **Knowledge cutoff:** Oct 2024 (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/helicone/models/llama-prompt-guard-2-22m.toml)
- **Input:** Text (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/helicone/models/llama-prompt-guard-2-22m.toml)
- **Output:** Text (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/helicone/models/llama-prompt-guard-2-22m.toml)
- **Reasoning mode:** No (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/helicone/models/llama-prompt-guard-2-22m.toml)
- **Tool calling:** No (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/helicone/models/llama-prompt-guard-2-22m.toml)

## API pricing (USD per 1M tokens)

| Provider | Input | Output | Cached input | Context |
|---|---|---|---|---|
| Groq | $0.03 | $0.03 | – | 512 |

## FAQ

### What is the context window of Meta Llama Prompt Guard 2 22M?

Meta Llama Prompt Guard 2 22M has a context window of 512 tokens (512) and can generate up to 2 tokens in a single response, according to models.dev.

### How much does the Meta Llama Prompt Guard 2 22M API cost?

Meta Llama Prompt Guard 2 22M is listed from $0.03 per million input tokens and $0.03 per million output tokens (Groq). Prices last verified 22 Oct 2026.

### Is Meta Llama Prompt Guard 2 22M open source?

Meta Llama Prompt Guard 2 22M is an open-weight model. Check the license terms before commercial use.

### When was Meta Llama Prompt Guard 2 22M released?

Meta released Meta Llama Prompt Guard 2 22M on 29 Apr 2025, according to models.dev.

---
Data licensed CC BY 4.0. Cite as: AI Model Directory, "Meta Llama Prompt Guard 2 22M", https://www.aimodel.directory/models/llama-prompt-guard-2-22m