# Hugging Face data API for AI agents: from $0.001 per call ($1 per 1,000), no subscription

> Whizbrew's Hugging Face data API costs from $0.001 per call ($1 per 1,000 calls); a typical call is $0.001 ($1 per 1,000). Only a call that returns data is charged: an empty or failed call is free. There is no subscription and no monthly fee. A new account starts with $2 of free credit, with no card.

It reads public Hugging Face data: 5 tools, live at the moment of the call, over one API key or one MCP server (`https://whizbrew.com/mcp`). It is an API, not a scraper you run: no proxies, no browser and no scraping code on your side. The same key reaches 893 tools for other platforms ([all APIs](https://whizbrew.com/api), [all tools](https://whizbrew.com/tools?a=11b18ab528ec), [pricing](https://whizbrew.com/pricing?a=11b18ab528ec)).

## Pricing

| Tool | What it returns | Per call that returns data | Per 1,000 calls |
|---|---|---|---|
| `huggingface.trending` | Trending AI models, datasets or spaces on Hugging Face, with likes, downloads and trending score. | $0.001 | $1 |
| `huggingface.model` | One Hugging Face model: downloads of the last month, likes, task, license and how many spaces use it. | $0.001 | $1 |
| `huggingface.models_search` | Search AI models on Hugging Face by words, author, task or tag, most downloaded first: id, author, downloads of the last month, likes, task, library, tags and creation date. One read of the public Hub API; no login, so private and gated-only details are not there. Nothing found is an empty list. | $0.001 | $1 |
| `huggingface.datasets_search` | Search datasets on Hugging Face by words, author or tag, most downloaded first: id, author, downloads of the last month, likes, description, tags and dates. One read of the public Hub API; no login. Nothing found is an empty list. | $0.001 | $1 |
| `huggingface.spaces_search` | Search Spaces (hosted AI demo apps) on Hugging Face by words, author or tag, most liked first: id, author, likes, the SDK it is built with, tags and creation date. One read of the public Hub API; no login. Spaces have no download count. Nothing found is an empty list. | $0.001 | $1 |

## Example calls

`huggingface.trending` ($0.001 per call): Trending AI models, datasets or spaces on Hugging Face, with likes, downloads and trending score

```
curl -s -X POST https://whizbrew.com/v1/call/huggingface.trending \
  -H "Authorization: Bearer $WHIZBREW_KEY" -H "content-type: application/json" \
  -d '{"limit":3}'
```

`huggingface.model` ($0.001 per call): One Hugging Face model: downloads of the last month, likes, task, license and how many spaces use it

```
curl -s -X POST https://whizbrew.com/v1/call/huggingface.model \
  -H "Authorization: Bearer $WHIZBREW_KEY" -H "content-type: application/json" \
  -d '{"id":"openai-community/gpt2"}'
```

`huggingface.models_search` ($0.001 per call): Search AI models on Hugging Face by words, author, task or tag, most downloaded first: id, author, downloads o

```
curl -s -X POST https://whizbrew.com/v1/call/huggingface.models_search \
  -H "Authorization: Bearer $WHIZBREW_KEY" -H "content-type: application/json" \
  -d '{"query":"whisper","limit":3}'
```

Over MCP the same call is `call_tool` with `{"tool": "<name>", "arguments": {…}}`. How often each tool works, measured from real calls: [https://whizbrew.com/status](https://whizbrew.com/status).

## Start

- **An agent on its own**: read [https://whizbrew.com/agents.md?a=11b18ab528ec](https://whizbrew.com/agents.md?src=body-text&a=11b18ab528ec). It asks for a free quote (`GET https://whizbrew.com/v1/agent/guide?goal=<the task in plain words>`), then one request (`POST https://whizbrew.com/v1/agent/start?a=11b18ab528ec`) gives it a key that works at once with the free credit, with no email and no sign-up form. Its human claims the account later.
- **A chat app** (ChatGPT, Claude, Gemini): add the MCP server `https://whizbrew.com/mcp` as a custom connector and sign in. No key to copy.
- **An agent with a shell** (Claude Code, Codex, Cursor): read [https://whizbrew.com/llms.txt?a=11b18ab528ec](https://whizbrew.com/llms.txt?src=body-text&a=11b18ab528ec). The agent starts a pairing with one request, its user approves with one click, and the agent holds its own key.
- **Plain HTTP**: create an account at [https://whizbrew.com/app#signup](https://whizbrew.com/app#signup), make a key, then `POST https://whizbrew.com/v1/call/<tool>` with `Authorization: Bearer <key>`.
