For developers
Hallox API
Coming Soon
Give your AI product a trust layer with one API call
One endpoint
POST to /v1/analyse and get a full trust breakdown. No complex setup, no SDKs required.
JSON response
Structured JSON with factual accuracy, hallucination risk, toxicity, bias, verdict, and per-claim corrections.
Any AI model
Works with outputs from GPT-4, Claude, Gemini, Llama, or any text-based AI system.
analyse.jsPREVIEW
// Analyse any AI response in one callconst response = await fetch('https://api.hallox.ai/v1/analyse', { method: 'POST', headers: { 'Authorization': `Bearer ${API_KEY}`, 'Content-Type': 'application/json', }, body: JSON.stringify({ text: 'Paste the AI response to fact-check here...', model: 'gpt-4o', }),});const { overallTrust, verdict, corrections } = await response.json();// { overallTrust: 82, verdict: "CAUTION", corrections: [...] }API access will be available to waitlist members first.