Flux
Dashboard

Getting started

Create a key and make your first web search and Lineage requests in a few minutes, with curl or the OpenAI SDK.

Flux gives AI agents two APIs behind one key. Search returns ranked, live web results. Lineage returns the passages of a page or document that answer a question, with provenance written in: the people, places, organisations and dates they mention are named inline.

1. Create an account

Sign up. New accounts start with free credit.

2. Create an API key

Open API keys in the dashboard and create a key. Send it on every request as a bearer token.

bash
export FLUX_API_KEY="<your key>"

3. Run Lineage on a page

Send a URL, or the text itself, and the question the passages should answer.

bash
curl https://fluxsearch.io/api/v1/provenance \
  -H "Authorization: Bearer $FLUX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"url": "https://example.com/article", "query": "revenue"}'

Each result's enriched_text is the passage with its provenance inline. It is never null, so it can go straight to your model.

4. Search the web

bash
curl https://fluxsearch.io/api/v1/search \
  -H "Authorization: Bearer $FLUX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"query": "acme corp revenue", "max_results": 10, "enrich": 5}'

Results you ask to enrich are read and carry the same Lineage passages and provenance, in lineage on each result.

5. Or use the OpenAI SDK

Point the SDK's base URL at Flux. The model picks the product: flux-search-1 or flux-lineage-1.

python
import os
from openai import OpenAI
 
client = OpenAI(base_url="https://fluxsearch.io/api/v1", api_key=os.environ["FLUX_API_KEY"])
 
response = client.chat.completions.create(
    model="flux-search-1",
    messages=[{"role": "user", "content": "acme corp revenue"}],
)
print(response.choices[0].message.content)

Next steps