A grounded answer for about half a cent

Ask ask_pipeworx a question and it routes to the right live source, runs the call, and returns a cited answer. On credits that typically costs ~$0.005 — the request fee plus the actual inference, priced at what it cost us. The same grounded answer assembled through a web-search API stack runs $0.16–$1.20 per answer.

Methodology and per-question receipts: More accurate and 31×–228× less expensive — 15 questions, 11 ways (blind-graded).

Anonymous

For trying things out.

Free forever
  • 50 tool calls / day
  • All 5,433 tools across 1411 packs
  • No signup, no API keys
  • 24-hour memory TTL
  • Full MCP protocol (Tools + Resources + Prompts)
Recommended

Registered (Free)

For real prototypes.

Free with GitHub signup
  • 200 tool calls / day
  • Everything in Anonymous
  • Persistent remember / recall memory
  • Bearer token for stable cross-session identity
  • Usage dashboard at /account
  • Add credits to keep going past the daily cap

Paid

Past the free tiers, everything is credits (1 credit = $0.0001), priced by your monthly volume. Brackets are marginal, like tax brackets — crossing one only changes the price of the calls after it.

First 10,000 10,001 – 20,000 20,001 – 50,000 50,000+
Standard APIs 50 credits ($0.005) 30 credits ($0.003) 20 credits ($0.002) 10 credits ($0.001)
Premium APIs cost × 1.5 cost × 1.4 cost × 1.3 cost × 1.2

Example: 12,000 standard requests in a month = 10,000 × $0.005 + 2,000 × $0.003 = $56 — not 12,000 at either rate. Premium APIs are the few sources that bill us per unit (per email found, per page, per market); you pay their real cost times your bracket's markup — ×1.5 down to ×1.2, same volume counter as standard requests — floored at $0.01, itemized in your ledger. The counter resets at UTC month start; your current bracket is on your account page.

Pro — $99/month

50,000 requests a month included, then a tenth of a cent ($0.001) per request after that. The crossover is about 30,000 requests a month: below that pay-as-you-go is cheaper and you shouldn't pay for headroom you won't use; at the full 50,000 included, Pro is $99 against $140 on credits.

Premium APIs still bill at their real cost, at the best markup (×1.2) from your first call instead of after 50,000. Everything else works the same: no separate quota, same tools, cancel whenever.

1 · Requests

One tool call to any of the 5,433 tools. Volume-tiered by calendar month, priced marginally — your cheap band never re-prices your early calls, and heavy volume keeps the old $0.001 rate.

2 · Answers

ask_pipeworx-class tools run real inference to route and ground your question. An answer costs the measured inference at 1.5× what it costs us, plus one request for the underlying tool call. Cheap routing means a cheap answer — you pay what it actually took, not a flat guess.

3 · Metered sources

A few packs front vendors that bill us per unit — per email found, per page, per market. Those settle at the vendor's real unit cost times your bracket's markup (×1.5 down to ×1.2 by monthly volume), floored at $0.01, itemized in your ledger. Bring your own key for any of them and we charge nothing extra.

Bring your own keys for paid upstream sources

Some packs wrap upstream APIs that require their own paid keys — FRED, ATTOM, Alpha Vantage Premium, Altos, and others. Pipeworx doesn't mark those up. Pass your own key per call via the _apiKey argument and you pay the upstream provider directly. The Pipeworx tier still counts against your daily/billed limit.

For sources that are free upstream (SEC, FDA, FRED Basic, ClinicalTrials, RxNorm, USPTO, Census, BLS, OpenAlex, EPA, NOAA, and ~280 others), no _apiKey is needed — Pipeworx covers all the upstream traffic.

Read more →

Vertical bundles include managed paid sources

The Pipeworx Verticals (Housing, FinTech, Pharma, GovCon, Trade, Green, Agri) bundle packs by domain and include managed access to paid upstream sources relevant to that vertical — so subscribers don't need to bring their own keys for ATTOM, Altos, etc.

Vertical subscriptions are priced per-vertical and stack on top of your account tier. See individual vertical pages for current pricing.

What counts as a "request"?

One MCP tool invocation through the gateway — an actual question asked of an actual source. Setting up the connection is free: initialize, tools/list, resources/list and keepalive pings run on their own budget and never draw down your daily limit or your credits.

What happens when I hit my daily limit?

Anonymous callers get an HTTP 429 until UTC midnight. Signed-in accounts with credits keep going seamlessly at the request rates above; without credits you get a clear message with your options. Nothing is ever charged silently past a limit you didn't fund.

Do answers cost more than requests?

Only by what the inference actually cost. A routed answer on the standard model adds about a tenth of a cent; deeper modes (ask_pipeworx_grounded, deep_research) run more model calls and settle at their measured cost — itemized, never estimated in our favor.

What if a call fails?

Errors come back structured — a retry hint, the upstream status, and what to try instead — so your agent can route around the failure instead of burning calls on it. Every pack's live health is on /status, and persistent breakage reported via pipeworx_feedback is reviewed daily.

Can I cap my spend?

Yes — every account has a daily spend ceiling (it starts at $10/day and grows with account history, or set it yourself at /account), and auto-refill is off unless you turn it on.

Where do I see what I've used?

Signed-in accounts have a usage dashboard at /account showing calls, top packs, error rate, current-month volume and bracket, and the full credit ledger.