Docs
Early Access hello@polariapi.com
Getting Started

Quickstart

Make your first API call in under five minutes. This guide covers authentication, a basic request, and a walkthrough of what Polari returns.

Early access. Polari is in private beta. Request access at hello@polariapi.com — beta customers get 6 months free on the Professional tier.

1. Get your API key

After your early access request is approved, you'll receive a key via email. Keys follow the format:

FORMAT
pk_live_xxxxxxxxxxxxxxxxxxxxxxxxxxxx

Keep this key secret. Do not commit it to version control — use environment variables.

namespace
Your key carries a namespace mode. By default, keys read and write to the shared corpus — articles you submit contribute to the global pool and are quality-gated at 0.53. If you need your data isolated (proprietary content, bypassed quality gate), request a private namespace key when you sign up. See API keys & namespace mode for the full model.

2. Authenticate

Pass your key as a Bearer token in the Authorization header on every request.

BASH
curl https://api.polariapi.com/v1/clusters \ -H "Authorization: Bearer pk_live_your_key_here"

3. Fetch your first clusters

Clusters are the most immediately useful endpoint — they return cross-source story groups with entities, sentiment, and article lists already assembled.

PYTHON
import requests API_KEY = "pk_live_your_key_here" BASE = "https://api.polariapi.com/v1" headers = {"Authorization": f"Bearer {API_KEY}"} resp = requests.get( f"{BASE}/clusters", headers=headers, params={ "topic": "artificial intelligence", "min_sources": 3, "time_range": "24h", } ) for cluster in resp.json()["clusters"]: print(cluster["title"], f"({cluster['source_count']} sources)")

4. Understand the response

Every cluster response includes the story title, article count, source count, confidence score, primary entities, and sentiment distribution:

JSON
{ "clusters": [ { "id": "clus_9x3k2m8f", "title": "AI Regulation Debate Intensifies", "article_count": 47, "source_count": 12, "confidence": 0.94, "sentiment_distribution": { "positive": 0.35, "neutral": 0.45, "negative": 0.20 }, "primary_entities": ["ent_...", "ent_..."], "articles": ["art_...", "art_..."] } ] }

Explore the layers

Polari processes every article through four semantic layers, then enriches story clusters with public discourse signal through the Echo layer. Each layer builds on the previous and unlocks richer endpoints.