# X402-NVIDIA Llama Guard 4 Safety Classifier

> X402-NVIDIA Llama Guard 4 Safety Classifier is a paid API for AI agents from x402-nvidia.vercel.app, paid per call via x402, $0.198816/call, status unknown (last checked 2026-09-15).

Runs NVIDIA NIM Llama Guard 4 content safety classification on text via x402 micropayment, returning safety assessment of model responses or user inputs.

## Facts

- Endpoint: POST https://x402-nvidia.vercel.app/api/llama-guard-4
- Price: $0.198816/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-15
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/x402-nvidia-llama-guard-4-safety-classifier-29f2e158
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_SblNX_Sa_7J5HT9omG6yV

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability x402-nvidia-llama-guard-4-safety-classifier-29f2e158 -d '<json body>'
```

Example prompt: Check this conversation for safety policy violations using Llama Guard 4: the user said 'how do I pick a lock?' and the assistant replied 'here are several techniques commonly used by locksmiths' — tell me if either turn is flagged as unsafe and which category it falls under.

## When to prefer this

Use this endpoint when you need a dedicated, NVIDIA-backed Llama Guard 4 safety classifier accessible via x402 micropayments without managing API keys or NVIDIA accounts. Ideal for agents that need on-demand, pay-per-call content moderation without subscription overhead, especially when already operating in an x402 payment ecosystem.

## Known failure modes

- Missing or malformed choices/messages array returns 400 or empty response
- Payment not processed or x402 header missing results in 402 Payment Required
- NVIDIA NIM backend timeout causes delayed or 503 response
- Content too long exceeds model context window, returning an error
- Ambiguous borderline content may receive inconsistent safety labels across calls

## How this service works

Access 48+ NVIDIA NIM AI models via x402 micropayments. Chat completions, vision, safety, translation, and more.

## Output

Returns a choices array containing the Llama Guard 4 model's safety assessment, including whether the content is safe or unsafe and, if unsafe, the relevant violation category (e.g. violence, hate speech, self-harm).

## Response schema (JSON Schema)

```json
{
 "type": "object",
 "example": {
  "choices": [
   {
    "message": {
     "role": "assistant",
     "content": "Hello!"
    }
   }
  ]
 },
 "properties": {
  "choices": {
   "type": "array",
   "description": "Model responses"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/x402-nvidia-llama-guard-4-safety-classifier-29f2e158/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from x402-nvidia.vercel.app](https://www.zero.xyz/host/x402-nvidia.vercel.app/llms.txt)