Parallel

Parallel turbo parallel/turbo

Parallel turbo, ~200ms. Available through Mux with the same request and response shape as every other model, at Parallel’s list price.

Call it

curl https://api.getmux.io/v1/search \
  -H "Authorization: Bearer $MUX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"query": "open source vector databases", "model": "parallel/turbo", "max_results": 10}'

What a request costs

Each request includes 10 results. Results beyond 10 cost $0.001 each, billed on what the provider returns and never on more than you asked for. Requests that fail are refunded.

Use it with a fallback

Put parallel/turbo first and a model from another provider after it. If Parallel errors or times out, Mux refunds the attempt and tries the next one. You pay for the model that answered.

const res = await mux.search({
  query: "open source vector databases",
  models: ["parallel/turbo","exa/instant"], // tried in order
});
console.log(res.model, res.attempts);

How Parallel behaves through Mux

Full content

Excerpts joined together

Domain filters

Applied natively, and re-checked by Mux

published_before

Not applied upstream (Mux filters afterwards)

provider_options

max_chars_totalclient_modeladvanced_settings.excerpt_settingsadvanced_settings.location

Details in How providers are normalized. Parallel’s own documentation is at parallel.ai.