EmailContact
Skip to main content

GPTfake API

Professional REST API for accessing AI censorship monitoring data, bias metrics, and research insights.

Overview​

Our API provides programmatic access to:

  • Real-time censorship metrics across AI models
  • Historical trend data and policy changes
  • Bias detection scores and analysis
  • Research publications and datasets

Quick Start​

Authentication​

All API requests require an API key:

curl -H "Authorization: Bearer YOUR_API_KEY" \
  https://api.gptfake.com/v1/monitoring/chatgpt/metrics

Base URL​

https://api.gptfake.com/v1

Core Endpoints​

Monitoring Data​

Get current censorship metrics for any monitored model:

GET /monitoring/{model}/metrics

Models: chatgpt, claude, gemini, mistral, qwen

Example:

curl https://api.gptfake.com/v1/monitoring/chatgpt/metrics \
  -H "Authorization: Bearer YOUR_API_KEY"

Response:

{
  "model": "chatgpt",
  "version": "gpt-4o",
  "metrics": {
    "overall_refusal_rate": 18.7,
    "political_refusal_rate": 34.2,
    "safety_refusal_rate": 67.8,
    "bias_score": -12
  },
  "trend": "rising",
  "last_updated": "2024-11-25T00:00:00Z"
}

Historical Data​

Get historical censorship trends:

GET /monitoring/{model}/history?period=30d

Parameters:

  • period: 7d, 30d, 90d, 1y
  • category: political, safety, social, all

Model Comparison​

Compare metrics across models:

GET /monitoring/compare?models=chatgpt,claude,gemini

Bias Analysis​

Get detailed bias analysis:

GET /bias/{model}?topics=political,social,economic

Research Endpoints​

Publications​

Get published research:

GET /research/publications?year=2024

Datasets​

Access research datasets:

GET /research/datasets
GET /research/datasets/{dataset_id}/download

Webhooks​

Get notified of significant changes:

POST /webhooks
{
  "url": "https://your-app.com/webhook",
  "events": ["policy_change", "censorship_spike"],
  "models": ["chatgpt", "claude"]
}

Rate Limits​

TierRequests/Hour
Free100
Researcher1,000
Pro10,000
EnterpriseCustom

Error Handling​

{
  "error": {
    "code": "RATE_LIMIT_EXCEEDED",
    "message": "Rate limit exceeded",
    "retry_after": 3600
  }
}

Error Codes​

  • 400 - Bad Request
  • 401 - Unauthorized
  • 403 - Forbidden
  • 404 - Model/Resource Not Found
  • 429 - Rate Limit Exceeded

SDKs​

Python​

pip install gptfake-sdk
from gptfake import GPTfakeClient

client = GPTfakeClient('YOUR_API_KEY')

# Get metrics
metrics = client.get_metrics('chatgpt')
print(f"Refusal rate: {metrics.overall_refusal_rate}%")

# Get history
history = client.get_history('chatgpt', period='30d')

JavaScript​

npm install @gptfake/sdk
import { GPTfakeClient } from '@gptfake/sdk';

const client = new GPTfakeClient('YOUR_API_KEY');

// Get metrics
const metrics = await client.getMetrics('chatgpt');
console.log(`Refusal rate: ${metrics.overallRefusalRate}%`);

Support​

Next Steps​

Last updated on