Parallel Decisions

Context

32K

Price

~~$0.001 / call~~Free

Input → output

text → typed answers

Weights

–

Providers

1 provider · price, latency and uptime

ProviderPer callLatency p50Uptime 24hRetention

Parallel

~~$0.001~~Free0.6 s – may retain

Listed price: $0.001 / request
Charged per call: ~~$0.001~~Free
Precision: –
Regions: default
Headquarters: –
Data retention: May retain prompts

By default the router tries our listings in routing order and fails over before the first byte — never mid-answer. Uptime is our own 24h measurement; a dash means too few requests to measure yet. Open the row for the provider's specs and policies.

Parallel: Parallel answers each call from the web: it searches for evidence on your state and questions, and the answer lists up to 10 of its sources as citations. The list price is $0.001 a call for 1 to 100 questions, Parallel's own with no markup, and a call whose answers are all refusals counts as a call; the price card above shows what a call is charged now.

Parallel keeps what it is sent. Under its standard customer terms it may use the state, the questions and the answers, not linked to any individual, to train and improve its models; zero data retention is on its Enterprise plan only, and our account is not on it. So the listing says it retains prompts. See the data policy.

Quickstart

plain HTTP — one JSON call, one typed answer per question

curl https://api.routerplus.com/v1/decisions \
  -H "Authorization: Bearer $TM_API_KEY" -H "content-type: application/json" \
  -d '{"model":"parallel/decisions",
       "state":"Help! My payouts have been failing for 3 days.",
       "questions":{
         "team":{"type":"choice","instructions":"Which team should handle this?",
                 "criteria":{"billing":"Payments and refunds","technical":"Bugs and outages"}},
         "urgent":{"type":"noul","instructions":"Is this urgent?"}}}'
# answers.team.choice, answers.team.probabilities, answers.urgent.noul

import os, httpx
r = httpx.post("https://api.routerplus.com/v1/decisions",
  headers={"Authorization": f"Bearer {os.environ['TM_API_KEY']}"},
  json={"model": "parallel/decisions", "state": "Help! My payouts have been failing for 3 days.",
        "questions": {"urgent": {"type": "noul", "instructions": "Is this urgent?"}}})
r.json()["answers"]["urgent"]["noul"]  # probability of yes

const r = await fetch("https://api.routerplus.com/v1/decisions", {
  method: "POST",
  headers: { authorization: `Bearer ${process.env.TM_API_KEY}`, "content-type": "application/json" },
  body: JSON.stringify({ model: "parallel/decisions", state: "Help! My payouts have been failing for 3 days.",
    questions: { urgent: { type: "noul", instructions: "Is this urgent?" } } }),
});
const { answers } = await r.json();  // answers.urgent.noul: probability of yes

.qs pre[hidden] { display: block; border-top: 1px solid var(--code-line); } .qsb { display: none; } .qcode::before { content: attr(data-lang); display: block; color: var(--code-muted); font-size: 11px; margin-bottom: 6px; }

Pricing

per call of 1-100 questions: Parallel's list price, no markup · every answered call is billed, one whose answers are all refusals included · free while the offer lasts: no decision is charged, and usage.cost is 0

request

~~$0.001 / request · 1-100 questions~~Free

Every billed response carries usage.cost — recompute your bill from the wire.