Comparison
PDFops vs Anvil
Anvil is the most mature PDF-filling API on the market — a seven-year-old, venture-backed document-automation suite with AI field-detection, e-signature, and no-code workflows. PDFops is a deliberately narrow primitive: fill AcroForm fields and merge PDFs, deterministically, on the edge. This page is the honest split — where Anvil's breadth wins, and where a thin edge-native tool wins on substrate and price.
The honest framing
If you need to detect fields on flat or scanned PDFs, collect signatures, build multi-step webform-to-PDF-to-sign workflows, or lean on official SDKs and a polished dashboard — Anvil is the better product, and it isn’t close. It has years of surface area PDFops does not try to match.
If you already have AcroForm templates (or can add fields once), and what you actually need is data-in / filled-PDF-out at low latency and low cost, the calculus flips. Anvil runs every call through its hosted cloud and meters each programmatic fill at $0.10 per document. PDFops runs on the V8 edge with no headless browser, fills deterministically, merges in the same primitive, and is designed to cost roughly $0.004 per call — about 25× less per fill at standard rates.
One structural difference worth naming. Anvil is, by design, a cloud middleman: every fill is a round-trip to app.useanvil.com, with no edge or self-host option. PDFops is edge-hosted today, and an embeddable, in-function library — the version where the fill runs inside your own Worker with no third-party round-trip — is the next step on the roadmap.
Side-by-side
| PDFops | Anvil | |
|---|---|---|
| Primitive | AcroForm fill + merge (deterministic) | Full document-automation suite (fill, generate, e-sign, workflows) |
| Per-fill price | Free during beta; ~$0.004/call planned | $0.10/PDF (bulk discounts at high volume) |
| Paid plans | Indie ~$16/mo (4k calls) planned | AI Pack $99/mo, Product Pack $425/mo |
| Hosting substrate | Vercel Edge (V8 isolates), no headless browser | Anvil cloud only (no edge, no self-host) |
| Determinism | Deterministic, audit-safe by construction | Deterministic fill + optional AI detection |
| PDF merge | Yes (core endpoint) | Yes |
| Flat / scanned field detection | No — requires AcroForm fields | Yes — Document AI (computer vision) |
| E-signature | No | Yes (Etch) |
| No-code workflows / webforms | No | Yes |
| Official SDKs | Not yet (HTTP API) | Node, Python, .NET + React UI |
| Free tier | 100 req/IP/month, no signup | Dashboard only — metered API needs a card |
| Maturity | MVP / beta | 7 years, ~$15M raised, established |
| Best for | Template fill + merge at scale on edge stacks, cost- and latency-sensitive | Teams needing detection, e-sign, workflows, or a full platform |
Anvil pricing verified at useanvil.com/pricing as of 2026-06-09. PDFops post-beta pricing is planned, not final.
If the narrow tool fits, try it in 30 seconds
Drop a form PDF into the Form-Field Inspector to see its AcroForm fields and the exact call. The fill shape is one HTTP request:
curl -X POST https://pdfops.dev/api/fill-form \
-F "pdf=@template.pdf" \
-F 'fields={"customer_name":"Acme Co","total":"$1,250.00"}' \
-o filled.pdfNo API key, no signup during beta. The endpoint reference is in the fill-form docs, and the deterministic-vs-AI argument is in this essay.
Frequently asked
Is PDFops a drop-in replacement for Anvil?
No — Anvil is a broad platform; PDFops is one narrow primitive. Anvil does flat-PDF field detection, e-sign, and workflows; PDFops fills AcroForm fields and merges, deterministically, on the edge. If you need the platform features, Anvil. If you have templates and want edge latency, determinism, and ~25× lower per-call cost, PDFops.
How much cheaper is PDFops per call?
Anvil meters fills at $0.10/PDF on its paid plans ($99–$425/mo), with high-volume bulk discounts. PDFops is free during beta and planned at ~$0.004/call (~$16 for 4,000) — roughly 25× cheaper at standard rates, because the V8-edge substrate with no headless browser is dramatically cheaper to run. Verified at useanvil.com/pricing on 2026-06-09.
Does PDFops detect fields on flat or scanned PDFs?
No — this is Anvil's biggest edge and an honest gap. Anvil's Document AI tags fields on flat PDFs and scans via computer vision; PDFops needs the PDF to already have AcroForm fields. The free Inspector shows the fields a PDF has; if it has none, you add them once in Acrobat, Preview, LibreOffice Draw, or pdftk. One-time for recurring templates; Anvil wins for arbitrary un-tagged PDFs at scale.
Does PDFops do e-signature or workflows?
No. Anvil has Etch e-sign, Webforms, and no-code Workflows; PDFops deliberately doesn't. PDFops is the fill + merge primitive, not a platform. If e-sign or workflows are core, Anvil (or a dedicated e-sign API) is the right pick.
Where does the PDFops call run in production?
Anywhere — it's an HTTP API. Your calling code runs on Cloudflare Workers, Vercel Edge, Deno Deploy, Bun, Lambda, or a Node backend; PDFops itself is edge-hosted for low latency. The in-function embeddable library (fill runs inside your own Worker, no round-trip) is on the roadmap.
Move when ready
If the deterministic fill + merge primitive matches your usage, join the waitlist and tell me your volume and the forms you fill most — that signal is what pricing tiers and the in-function library get built around.