Search

How to get your Parallel API usage, balance and spend programmatically

Parallel has no public balance or usage endpoint. Billing is prepaid in the dashboard at platform.parallel.ai, and responses don't report cost.

What the Parallel API exposes

SignalParallel
BalanceNo
Spend historyNo
Cost breakdownNo
Plan and quotaNo
Cost on each responseNo
Rate-limit headersNone
CredentialAPI key

No billing API, so measure from your side

Argus prices each Parallel call from what the SDK records: the endpoint and mode, priced at Parallel's per-request rates (Search $1–$5 per 1,000, the Task API by processor tier).

Credential. A Parallel API key, sent as x-api-key. Argus validates it with a turbo search, the cheapest mode.

Gotchas

  • Only POST requests that create work count against the rate limit; polling a task's status or result doesn't.
  • Task API cost varies by processor tier by orders of magnitude. Record which processor each call used.

How Parallel charges

Usage-priced per request: Search $1–$5/1k, Task API by processor tier ($5–$2,400/1k tasks). Prepaid/dashboard billing; no public balance/usage API.

At zero. Prepaid/dashboard billing (platform.parallel.ai); no public balance/usage API. Qualified startups may receive up to $250 free credits.

  • 600 requests per minute, per key. Only POST (creating) requests count; status/result polls don't.
  • 2,000 requests per minute, per key.

Or let Argus track it

Paste the key once. Argus keeps the ledger, and your coding agent reads it through the MCP server.

  • With the SDK, itemizes each call by endpoint and prices it at Parallel's per-operation rates.
  • Emails you on low balance, days to zero, a spend spike, or projected overage.
  • Hands all of it to your coding agent over MCP, so it can ask what a run cost or how much runway is left without leaving the editor.
> what has parallel cost today, and how long until it runs out?

get_overview       → parallel: mtd spend, burn per day, days to zero
get_request_costs  { "provider": "parallel", "limit": 25 }
get_advisories     → credits_low, spend_spike, rate_limit_low …

Questions

Does Parallel have an API for its balance?
Parallel has no public balance or usage endpoint. Billing is prepaid in the dashboard at platform.parallel.ai, and responses don't report cost.
Can I get Parallel spend per day from the API?
No. Count calls and price them yourself, or let the Argus SDK do it.
What happens when Parallel credits run out?
Prepaid/dashboard billing (platform.parallel.ai); no public balance/usage API. Qualified startups may receive up to $250 free credits.
What key does Argus need for Parallel?
A Parallel API key, sent as x-api-key. Argus validates it with a turbo search, the cheapest mode.

Related

Pricing and limits verified 2026-07-21 against docs.parallel.ai/getting-started/pricing, docs.parallel.ai/getting-started/rate-limits, docs.parallel.ai/search/search-quickstart. Endpoints are the ones the Argus Parallel connector calls.

Let your coding agent watch the meter.

Connect a provider key, point Claude Code or Cursor at the Argus MCP server, and ask what the last run cost.