LIVE
OPUS 4.7$15 / $75per Mtok
SONNET 4.6$3 / $15per Mtok
GPT-5.5$10 / $30per Mtok
GEMINI 3.1$3.50 / $10.50per Mtok
SWE-BENCHleader Claude Opus 4.772.1%
MMLU-PROleader Opus 4.788.4
VALS FINANCEleader Opus 4.764.4%
AFTAv1.0 whitepaper live at /whitepaper
OPUS 4.7$15 / $75per Mtok
SONNET 4.6$3 / $15per Mtok
GPT-5.5$10 / $30per Mtok
GEMINI 3.1$3.50 / $10.50per Mtok
SWE-BENCHleader Claude Opus 4.772.1%
MMLU-PROleader Opus 4.788.4
VALS FINANCEleader Opus 4.764.4%
AFTAv1.0 whitepaper live at /whitepaper
All systems operational0 AI providers monitored, polled every 2 minutes
Live status
All endpoints

Premium Verified News Feed

1 credit
GET /api/premium/history/news/verified

The /api/premium/history/news/verified endpoint is the verified feed: cross-source story clusters filtered to those that cleared a corroboration threshold. Default min_sources=4 returns the corroboration_band="broad" subset. Each cluster carries source_count, sources list, article_ids, hero article (earliest publishedAt), and a corroboration_band tag. Embedding-based clustering at cosine threshold 0.82 across 12+ AI-relevant news sources.

When to use this endpoint

When an agent should NOT act on a single source. Finance agents reading news for trade signals, research agents pulling stories for synthesis, briefing agents preparing reports. The agent specifies the minimum independent-source threshold and gets only stories that cleared it. Single-day or 30-day range.

Parameters

NameInTypeDescription
datequerystringSingle UTC date (YYYY-MM-DD); pair with from/to for rangese.g. 2026-05-09
fromquerystringRange start (YYYY-MM-DD), max 30-day windowe.g. 2026-05-01
toquerystringRange end (YYYY-MM-DD)e.g. 2026-05-09
min_sourcesquerynumberMinimum corroborating sources (2-50, default 4)e.g. 4

* required

Example response

{
  "ok": true,
  "date": "2026-05-09",
  "min_sources_filter": 4,
  "items": [
    {
      "cluster_id": "k3mn8q",
      "date": "2026-05-09",
      "source_count": 5,
      "corroboration_band": "broad",
      "sources": ["anthropic.com", "techcrunch.com", "theverge.com", "reuters.com", "bloomberg.com"],
      "hero": { "title": "Anthropic Ships Mythos to Defenders First", "url": "https://www.anthropic.com/news/mythos", "source": "Anthropic Blog", "publishedAt": "2026-05-07T18:30:00Z" },
      "first_seen_at": "2026-05-07T18:30:00Z"
    }
  ],
  "billing": { "credits_charged": 1, "credits_remaining": 49 }
}

Code samples

Python SDK

from tensorfeed import TensorFeed

tf = TensorFeed(token="tf_live_...")
feed = tf._get("/premium/history/news/verified", date="2026-05-09", min_sources=4)
for cluster in feed["items"]:
    print(f"[{cluster['source_count']} sources] {cluster['hero']['title']}")

TypeScript SDK

const res = await fetch(
  "https://tensorfeed.ai/api/premium/history/news/verified?date=2026-05-09&min_sources=4",
  { headers: { Authorization: "Bearer tf_live_..." } }
);
const feed = await res.json();
console.log(`${feed.items.length} broadly corroborated stories`);

FAQ

How many sources are typically needed for "broad"?

Default min_sources=4 maps to corroboration_band="broad". Lower values return broader coverage at the cost of weaker corroboration. For highest-trust signals on important stories, 4 is the right floor; for emerging stories, 2 catches early multi-source pickup.

What is the news corpus?

12+ AI-relevant news sources polled hourly: Anthropic Blog, OpenAI Blog, Google AI Blog, Meta AI, HuggingFace, TechCrunch AI, The Verge AI, Ars Technica, VentureBeat AI, NVIDIA AI, ZDNet AI, Hacker News (AI-filtered).

Is this a fact-check?

No. We verify multiple sources reported the same story, not that the underlying claim is true. Five outlets repeating a misleading press release will all cluster together. Treat corroboration as a necessary but not sufficient signal.