# PQS: Prompt Quality Score – Cross-Model Comparison

> PQS: Prompt Quality Score – Cross-Model Comparison is a paid API for AI agents from api.relai.fi, paid per call via x402, $1.25/call, status unknown (last checked 2026-09-14).

Scores a prompt by running it against Claude Sonnet 4 and GPT-4o in parallel, then uses a third model to judge and compare the output quality

## Facts

- Endpoint: GET https://api.relai.fi/relay/1779373065272/api/score/compare
- Price: $1.25/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-14
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/pqs-prompt-quality-score-cross-model-comparison-02ad9624
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_PB3Qa6oWGTGT3lB0rkdad

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 pqs-prompt-quality-score-cross-model-comparison-02ad9624
```

Example prompt: Can you score the quality of this prompt using PQS — run it against both Claude Sonnet 4 and GPT-4o and tell me which model handles it better, using the 'software' vertical: 'You are a senior engineer. Review the following Python function for correctness and suggest improvements.'

## When to prefer this

Choose this endpoint when you need an objective, third-party judgment of prompt quality across multiple leading LLMs simultaneously. It is ideal for prompt engineers, AI developers, and researchers who want to benchmark their prompts against both Claude Sonnet 4 and GPT-4o rather than testing each model manually. Prefer it over single-model evaluation when cross-model comparison or domain-specific scoring (software, research, crypto, etc.) is required.

## Known failure modes

- Prompt text missing or empty — returns 400 validation error
- Prompt exceeds 10,000 character limit — returns 400
- Invalid vertical enum value — returns 400 schema validation error
- Payment not included or insufficient — returns 402 Payment Required
- Upstream model API timeout — may return 503 or partial result
- Judge model failure — scoring may be incomplete or unavailable

## How this service works

Cross-model scoring: Claude Sonnet 4 vs GPT-4o, judged by a third model

## Output

A structured quality score comparing how Claude Sonnet 4 and GPT-4o responded to the given prompt, judged by a third model, with per-model ratings and an overall verdict on which model performed better in the specified domain vertical.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "prompt": {
   "type": "string",
   "maxLength": 10000,
   "minLength": 1,
   "description": "Prompt to run through both Claude and GPT-4o for cross-model comparison"
  },
  "vertical": {
   "enum": [
    "software",
    "content",
    "business",
    "education",
    "science",
    "crypto",
    "general",
    "research"
   ],
   "type": "string",
   "default": "general",
   "description": "Domain: software/content/business/education/science/crypto/general/research"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/pqs-prompt-quality-score-cross-model-comparison-02ad9624/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from api.relai.fi](https://www.zero.xyz/host/api.relai.fi/llms.txt)