Pricing Series

1 credit

GET /api/premium/history/pricing/series

The pricing series endpoint returns the daily price points for one model across a date range, computed from the daily history snapshots. Each point includes input price, output price, and blended (average). The summary block surfaces first/latest, min/max, percent delta, and a count of detected price changes in the window.

When to use this endpoint

When you need to know how a model's price has moved over time. For a real-time alert when a price crosses a threshold, use a price watch instead.

Parameters

Name	In	Type	Description
model*	query	string	Model id or display name
from	query	string	Start date YYYY-MM-DD UTC (default 30 days ago)
to	query	string	End date YYYY-MM-DD UTC (default today)

* required

Example response

{
  "ok": true,
  "model": "Claude Opus 4.7",
  "provider": "Anthropic",
  "range": { "from": "2026-04-01", "to": "2026-04-27", "days": 27 },
  "points": [
    { "date": "2026-04-01", "input": 18, "output": 90, "blended": 54 },
    { "date": "2026-04-27", "input": 12, "output": 60, "blended": 36 }
  ],
  "summary": {
    "delta_pct_blended": -33.33,
    "changes_detected": 2,
    "min_blended": 36, "max_blended": 54
  }
}

Code samples

Python SDK

from tensorfeed import TensorFeed

tf = TensorFeed(token="tf_live_...")
series = tf.pricing_series(model="Claude Opus 4.7")
print(f"Price moved {series['summary']['delta_pct_blended']}%")

TypeScript SDK

import { TensorFeed } from 'tensorfeed';

const tf = new TensorFeed({ token: 'tf_live_...' });
const s = await tf.pricingSeries({ model: 'Claude Opus 4.7' });

MCP tool

Available via the TensorFeed MCP server as pricing_series. Add npx -y @tensorfeed/mcp-server to your Claude Desktop or Claude Code MCP config.

FAQ

How far back can I query?

Up to 90 days. The default window is 30 days back from today. Snapshots before our Phase 0 capture started are not available; the dataset compounds as we go.

What does "changes_detected" count?

The number of distinct days within the window where the input or output price changed from the previous day. A model whose price moved twice in 30 days reports changes_detected: 2.