POST
/api/context/load
~8 credits/1K tokens (new), free (cached)Smart context loader
Automatically detects input type (URL, batch URLs, or search query) and
returns compressed high-quality cached content (HQCC). Checks global cache
first, fetches and compresses on miss.
curl -s -X POST "$BASE/api/context/load" \
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-d '{"input": "https://example.com"}'Request Body
| Field | Type | Req | Default | Description |
|---|---|---|---|---|
| input | string | Y | — | |
| inputType | string (auto | url | urls | query) | N | auto | Force input type detection |
| processUncached | boolean | N | false | Process uncached URLs in batch mode |
| search | object | N | — | |
| └ topK | integer | N | 15 | Search results limit |
| processing | object | N | — | |
| └ maxConcurrent | integer | N | 3 | Max parallel compressions |
| └ strategy | string (auto | fast | quality) | N | auto | |
| return | object | N | — | |
| └ topK | integer | N | 5 | Return results limit |
| └ format | string (hqcc | raw | both) | N | hqcc | Response format (both returns hqcc + raw) |
| ranking | object | N | — | |
| └ preset | string (cache_first | relevance_first | balanced) | N | — |
Request Example
{
"input": "https://www.figure.ai/news/helix"
}Response Example
{
"success": true,
"requestId": "load_k2x9f",
"mode": "single_url",
"result": {
"url": "https://www.figure.ai/news/helix",
"title": "Helix: Vision-Language Action Model",
"hqcc": "# Helix\n\n## Overview\n...",
"cached": true,
"cachedAt": "2026-01-07T10:30:00Z",
"meta": {
"strategy": "Technical Content",
"model": "gpt-oss-120b"
}
},
"cost": {
"credits": 0,
"cached": true
},
"processingTime": 45
}