# PQS Cross-Model Scoring: Claude vs GPT-4o Comparison

> PQS Cross-Model Scoring: Claude vs GPT-4o Comparison is a paid API for AI agents from pqs.onchainintel.net, paid per call via x402, $1.25/call, status unknown (last checked 2026-09-15).

Runs the same prompt through both Claude and GPT-4o and uses a third model as judge to score and compare the outputs

## Facts

- Endpoint: GET https://pqs.onchainintel.net/api/score/compare
- Price: $1.25/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-15
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/pqs-onchainintel-net-a370506f
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_d_4R_lMXHPVlOQNDVekyY

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-onchainintel-net-a370506f
```

Example prompt: Run my prompt 'Explain how garbage collection works in Python' through both Claude and GPT-4o and use a third model to judge which one gives the better answer — the domain is software.

## When to prefer this

Use this endpoint when you need an objective, third-model-judged comparison of Claude vs GPT-4o on a specific prompt — ideal for prompt engineering research, selecting the best model for a domain, or validating which LLM to use before committing to a production integration. Prefer this over single-model scoring endpoints when the goal is model selection rather than prompt quality alone.

## Known failure modes

- Missing or empty prompt returns validation error
- Invalid vertical enum value causes bad request
- One upstream model (Claude or GPT-4o) unavailable causes partial or failed comparison
- Prompt too long for one or both models causes truncation or error
- Payment not included or invalid x402 header causes 402 Payment Required

## How this service works

PQS cross-model scoring - same prompt through Claude and GPT-4o, judged by a third model

## Output

Returns a scored comparison of Claude and GPT-4o outputs for the same prompt, with a third model acting as judge to evaluate quality differences, scores, and which model performed better in the given vertical.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "prompt": {
   "type": "string",
   "maxLength": 10000,
   "minLength": 1,
   "description": "Prompt to score"
  },
  "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-onchainintel-net-a370506f/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from pqs.onchainintel.net](https://www.zero.xyz/host/pqs.onchainintel.net/llms.txt)