Answer a question with citations
exa.web.answerGet an LLM answer to a question informed by Exa search results. Fully compatible with OpenAI's chat completions endpoint.
- Price
- $0.005/callAt provider cost
- Success rate
- 99.8%Last 30 days · 269,669 calls
- Latency p50
- 1.9 sp95 4.4 s
- Last verified
- Aug 27, 2026Verified working
POSThttps://toolswitch.dev/v1/tools/exa.web.answerPreviewTry onlineComing soonProvider docs ↗
- Uses platform quota
- Synchronous response
- Reads public data
Parameters
response is {answer, citations: [{url, title, publishedDate, author, text?}], costDollars}. This is the one Exa route that runs a model over the results; for raw results use exa.web.search.
This tool has no parameters. See provider docs ↗ Provider docs
Code samples
Replace $TOOLSWITCH_KEY with the key you created in the console.
Preview
curl -X POST 'https://toolswitch.dev/v1/tools/exa.web.answer' \
-H "X-Toolswitch-Key: $TOOLSWITCH_KEY" \
-H "Content-Type: application/json" \
--data-raw '{"query":"What year was the HNSW algorithm paper published and by whom?"}'import os
import json
import requests
response = requests.request(
"POST",
"https://toolswitch.dev/v1/tools/exa.web.answer",
headers={"X-Toolswitch-Key": os.environ["TOOLSWITCH_KEY"]},
json=json.loads("{\"query\":\"What year was the HNSW algorithm paper published and by whom?\"}"),
)
response.raise_for_status()
print(response.text)const response = await fetch("https://toolswitch.dev/v1/tools/exa.web.answer", {
method: "POST",
headers: {
"X-Toolswitch-Key": process.env.TOOLSWITCH_KEY ?? "",
"Content-Type": "application/json",
},
body: JSON.stringify(JSON.parse("{\"query\":\"What year was the HNSW algorithm paper published and by whom?\"}")),
});
if (!response.ok) throw new Error(`Request failed: ${response.status}`);
console.log(await response.text());This gateway API is in preview; its URL and authentication may change. The example parameters are sample data. Calls are billed at $0.005/call, including the service fee.