# ZeroGPU ZeroClick Content Moderation

> ZeroGPU ZeroClick Content Moderation is a paid API for AI agents from agents.zerogpu.ai, paid per call via MPP, price unknown, status unknown (last checked 2026-10-16).

Classifies text input against hate, sexual, violence, harassment, self-harm, and illicit content categories using a ZeroClick HMAC-authenticated OpenAI-compatible moderation endpoint.

## Facts

- Endpoint: POST https://agents.zerogpu.ai/zeroclick/v1/moderations?utm_source=zero.xyz
- Price: price unknown
- Payment: MPP
- Status: unknown
- Last checked: 2026-10-16
- Activations on Zero: 0
- Provider: agents.zerogpu.ai
- Website: https://agents.zerogpu.ai
- Canonical page: https://www.zero.xyz/c/agents-zerogpu-ai-zerogpu-zeroclick-content-moderation-1fb7da7a
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_AiD-EBv9Wt895uzwUV2A0

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 agents-zerogpu-ai-zerogpu-zeroclick-content-moderation-1fb7da7a -d '<json body>'
```

Example prompt: Can you check this user comment for harmful content — hate speech, violence, harassment, or anything illicit — and tell me which categories it violates: 'You deserve to be hurt for your beliefs'?

## When to prefer this

Choose this endpoint when you need OpenAI-compatible content moderation with ZeroClick HMAC authentication (no per-call OAuth, just a shared secret), want access to ZeroGPU's zlm-v1-moderation-edge model at low cost, need detailed multi-category classification (hate, sexual, violence, harassment, self-harm, illicit and subcategories), or are already integrated into the ZeroGPU agent ecosystem and want a drop-in replacement for OpenAI's omni-moderation-latest with token usage reporting.

## Known failure modes

- Invalid or missing zc-signature HMAC causes authentication failure
- Empty input string or empty input array returns validation error
- Unsupported model id not mappable to configured model causes model resolution error
- Malformed content parts object in multimodal input causes schema validation error
- Upstream ZeroGPU inference unavailability causes 5xx error
- Rate limit or quota exhaustion returns 429 or payment-required response

## How this service works

OpenAI-style Moderations endpoint. `model` may be omitted (the configured
moderation model is used) or given as an OpenAI id such as
`omni-moderation-latest`, which is mapped to it. `zc-usage` reports the
input and output tokens from the response's `usage`.

Authenticated by verifying the `zc-signature` HMAC over the raw
request bytes; the same orchestration as `/v1/moderations` runs behind it.


## Output

Returns a moderation object with a unique request id, the model used, and a results array — one entry per input. Each result includes a top-level `flagged` boolean, a `categories` object with boolean flags for hate, sexual, violence, harassment, self-harm, illicit, and their subcategories (e.g. sexual/minors, illicit/violent, hate/threatening, self-harm/intent, violence/graphic), plus token usage via the `zc-usage` header.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "required": [
  "input"
 ],
 "properties": {
  "input": {
   "oneOf": [
    {
     "type": "string",
     "minLength": 1
    },
    {
     "type": "array",
     "items": {
      "type": "string"
     },
     "minItems": 1
    },
    {
     "type": "array",
     "items": {
      "type": "object"
     },
     "minItems": 1
    }
   ],
   "description": "Text to classify. A single string, an array of strings (one result per element), or an array of content parts (`{ \"type\": \"text\", \"text\": \"...\" }`) forming one multi-modal input."
  },
  "model": {
   "type": "string",
   "default": "zlm-v1-moderation-edge",
   "example": "zlm-v1-moderation-edge",
   "description": "Moderation model to use. Defaults to the configured moderation model; OpenAI ids (`omni-moderation-latest`, `text-moderation-stable`) are accepted and mapped to it."
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "id": {
   "type": "string",
   "example": "modr-0a1b2c3d4e5f60718293a4b5c6d7e8f90",
   "description": "Unique identifier for the moderation request."
  },
  "model": {
   "type": "string",
   "example": "zlm-v1-moderation-edge",
   "description": "The model used for classification."
  },
  "results": {
   "type": "array",
   "items": {
    "type": "object",
    "properties": {
     "flagged": {
      "type": "boolean",
      "example": true,
      "description": "True when the model flagged the input in one or more categories."
     },
     "categories": {
      "type": "object",
      "properties": {
       "hate": {
        "type": "boolean",
        "example": false,
        "description": "Content that expresses, incites, or promotes hate based on a protected attribute."
       },
       "sexual": {
        "type": "boolean",
        "example": false,
        "description": "Content meant to arouse sexual excitement or that promotes sexual services."
       },
       "illicit": {
        "type": "boolean",
        "example": false,
        "description": "Content that gives advice or instruction on how to commit a wrongdoing."
       },
       "violence": {
        "type": "boolean",
        "example": true,
        "description": "Content that depicts death, violence, or physical injury."
       },
       "self-harm": {
        "type": "boolean",
        "example": false,
        "description": "Content that promotes, encourages, or depicts acts of self-harm."
       },
       "harassment": {
        "type": "boolean",
        "example": true,
        "description": "Content that expresses, incites, or promotes harassing language towards any target."
       },
       "sexual/minors": {
        "type": "boolean",
        "example": false,
        "description": "Sexual content that includes an individual under 18 years old."
       },
       "illicit/violent": {
        "type": "boolean",
        "example": false,
        "description": "Illicit content that also references violence or procuring a weapon."
       },
       "hate/threatening": {
        "type": "boolean",
        "example": false,
        "description": "Hateful content that also includes violence or serious harm towards a protected group."
       },
       "self-harm/intent": {
        "type": "boolean",
        "example": false,
        "description": "Content where the speaker expresses that they are engaging or intend to engage in self-harm."
… (truncated)
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/agents-zerogpu-ai-zerogpu-zeroclick-content-moderation-1fb7da7a/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from agents.zerogpu.ai](https://www.zero.xyz/host/agents.zerogpu.ai/llms.txt)