Do I need an account or API key?

No. Payment is the only identity: each request is paid individually via the x402 protocol, so there is nothing to sign up for and no key to manage. A standard x402 client library handles the pay-and-retry loop automatically.

How is pricing calculated?

Image endpoints are a flat price per call. PDF endpoints are priced per page of the input document — the server counts pages and quotes an exact amount in the 402 response, on both Base and Solana. The price is recomputed from the actual bytes on the paid retry, so a quote cannot be replayed against a different file, and you are never charged more than you paid.

Which chains and currency does it accept?

USDC on Base or Solana, settled through the PayAI x402 facilitator. The 402 response lists the exact amount required on each chain; pay whichever your wallet holds.

Every request is processed in memory and deleted the moment the response is sent — nothing is written to disk or logged. The only exception is the four Gemini-backed endpoints (image/ocr, pdf/ocr, pdf/tables, pdf/structured), which forward file content to Google for model-based reading. Those are clearly labeled and flagged in the discovery catalog; the deterministic extract and markdown paths never leave the server.

What happens if processing fails or times out?

You are not charged. Settlement happens after the work succeeds, so a request that errors, times out, or is rejected as unprocessable never takes payment.

How do agents discover the API?

Every endpoint, price, and input/output schema is published at https://api.pixo.tools/v1/discovery (x402 Bazaar format) and https://api.pixo.tools/v1/openapi.json (OpenAPI 3). The bare domain and https://api.pixo.tools/llms.txt also serve a machine-readable orientation.

x402 · HTTP 402USDC on Base + SolanaFiles never stored

The file API that agents pay per call — no account, no keys.

Compress, convert, resize, extract, and OCR images and PDFs over plain HTTP. Payment is per request via the x402 protocol — USDC on Base or Solana. An agent gets a price, pays, and gets the result. No sign-up, no API key, no subscription.

Base URLhttps://api.pixo.toolsDiscovery catalog

How payment works

One request, one micropayment

x402 turns HTTP's dormant 402 Payment Requiredstatus into a real payment step. No account to create, no key to manage — the agent's wallet is the identity.

Request

Your agent POSTs a file to an endpoint with no payment header.

Get a price

The server replies 402 with the exact price — for PDFs, computed from the page count — on both chains.

Pay

The wallet signs a USDC payment and retries the same request with an X-PAYMENT header.

Get the result

The server verifies, processes, settles on-chain, and returns the output plus a receipt.

Standard x402 client libraries (x402-fetch, the x402 SDKs for JS/Python/Rust) handle steps 2–4 automatically — your agent just makes the request. Pricing is tamper-proof in both directions: the server re-reads the file and re-computes the price on the paid retry, so a payment for a 1-page quote can never be replayed against a 300-page file — and you're never charged more than the quote you paid.

Endpoints

Twenty-one operations across images and PDFs

Every endpoint takes multipart/form-data with a file field. The in-process chip means the file is decoded in memory and deleted the moment the response is sent. → Gemini marks the document-AI endpoints, which forward content to Google for model-based reading.

Images

Flat price per call · both chains

Endpoint	Does	Privacy	Price
POST /v1/image/compress	Compress or convert (JPEG · PNG · WebP)	in-process	$0.005
POST /v1/image/resize	Resize — contain · cover · fill	in-process	$0.005
POST /v1/image/crop	Crop to an exact rectangle	in-process	$0.005
POST /v1/image/rotate	Rotate 90° · 180° · 270°	in-process	$0.005
POST /v1/image/flip	Mirror horizontally or vertically	in-process	$0.005
POST /v1/image/ocr	Extract text (JPEG · PNG · WebP — no GIF)	→ Gemini	$0.02

PDFs

Priced per page of the input · both chains

Endpoint	Does	Privacy	Price
POST /v1/pdf/compress	Shrink — lossless · balanced · max	in-process	$0.005 / page
POST /v1/pdf/split	Split into page ranges → ZIP	in-process	$0.005 / page
POST /v1/pdf/merge	Merge 2–20 PDFs → one	in-process	$0.005 / page
POST /v1/pdf/merge-compress	Merge then compress in one call — bundle	in-process	$0.008 / page
POST /v1/pdf/pages	Select · reorder · delete · rotate pages	in-process	$0.005 / page
POST /v1/pdf/protect	Password-protect with AES-256	in-process	$0.005 / page
POST /v1/pdf/unlock	Remove a password you supply — flat	in-process	$0.02 flat
POST /v1/pdf/extract	Text → JSON with a likelyScanned flag	in-process	$0.01 / page
POST /v1/pdf/images	Pull embedded images → ZIP of PNGs	in-process	$0.01 / page
POST /v1/pdf/markdown	Convert to Markdown for LLM / RAG	in-process	$0.02 / page
POST /v1/pdf/render	Render pages to PNG images → ZIP (36–300 dpi)	in-process	$0.01 / page
POST /v1/pdf/from-markdown	Turn Markdown into a PDF — flat	in-process	$0.01 flat
POST /v1/pdf/ocr	OCR a scanned PDF → text	→ Gemini	$0.03 / page
POST /v1/pdf/tables	Every table → structured JSON	→ Gemini	$0.03 / page
POST /v1/pdf/structured	Your JSON schema → typed JSON	→ Gemini	$0.04 / page

Free, no payment: GET /v1/health · /v1/openapi.json · /v1/discovery · /metrics. Every image endpoint also takes optional format/quality/colors fields, so one paid call can transform, convert, and compress together. Limits:files up to 50 MB; images up to 40 megapixels; merge up to 20 files / 95 MB total; the Gemini PDF endpoints (ocr/tables/structured) up to 30 pages; render up to 50 pages; structured schemas up to 16 KB.

The 402 handshake

What a request actually looks like

An unpaid request to a PDF endpoint returns the exact per-page price on both chains. A standard x402 client pays and retries automatically.

# Unpaid — ask a 3-page PDF what it costs
curl -X POST https://api.pixo.tools/v1/pdf/extract \
  -F "file=@report.pdf"

HTTP/1.1 402 Payment Required
// abridged — the real response also carries
// scheme, payTo, resource, and the USDC asset
{
  "x402Version": 1,
  "accepts": [
    { "scheme": "exact", "network": "base",
      "maxAmountRequired": "30000" },
    { "scheme": "exact", "network": "solana",
      "maxAmountRequired": "30000" }
  ]
}

↳ 3 pages × $0.01 = $0.03, quoted on Base and Solana.

Privacy

Files are processed in memory and deleted on response

The same promise as the pixo.tools web app, applied to the API: uploads live only in RAM-backed scratch space, are erased the instant the response is sent, and are never written to disk or logged.

In-process by default

Compression, conversion, resize, and every deterministic PDF text / Markdown / image operation runs entirely on our server. Content never leaves the box.

Document AI is the disclosed exception

Four endpoints — image/ocr, pdf/ocr, pdf/tables, pdf/structured — send content to Google Gemini for model-based reading. Each is labeled, flagged in discovery, and priced accordingly.

Nothing to sign up for

No account means no stored profile, no email, no history. The only identifier in a request is the payment itself.

Status codes

What your agent branches on

402	Payment required — the body lists the exact price on each chain. Pay and retry with X-PAYMENT.
400	Bad parameters — invalid option, too many merge files or split ranges, or a PDF over a page cap.
413	File too large (50 MB per file; 95 MB per merge request).
422	Unprocessable input — not a valid PDF/image, corrupt, or password-protected. Rejected before payment.
503	Document-AI not enabled on this deployment (the Gemini-backed endpoints only).
504	Processing timed out (60 s budget). Failed work is not charged.

Discovery

Built for agents to find on their own

Agents don't need this page — they read the service directly. Every endpoint, price, and input/output schema is published in machine-readable form.