How DocImprint compares to other document intelligence tools, and a breakdown of every extract mode.
Firecrawl returns Markdown chunks without cryptographic proof of capture. DocImprint returns tamper-evident evidence bundles with manifest SHA-256, EIP-191 signatures, free offline verify, and x402 USDC pay-per-call for agents.
LlamaParse returns structured JSON with optional citations but no tamper detection. DocImprint binds captures to SHA-256 manifests, supports legal hold, on-chain notarization, and Merkle citation proofs verifiable offline.
Use DocImprint when agents or auditors must prove what a document said at capture time — legal, finance, compliance, and research workflows. Use scrapers for bulk Markdown when proof is not required.
The core difference: only DocImprint produces cryptographic proof that a document has not been modified since capture.
| Firecrawl | LlamaParse | DocImprint | |
|---|---|---|---|
| Output | Markdown chunks | Structured JSON | Evidence bundle + hashes |
| Proof | None | Optional citations | Free offline verify |
| Agent pay | API keys | API keys | x402 USDC + API keys |
| Verify | — | — | Free offline + on-chain |
Firecrawl and LlamaParse are excellent tools for chunking and parsing. They do not produce cryptographic evidence or support agent-native x402 USDC payments.
Every mode is a single POST /v1/extract call. Mix lean and bundle per request.
| Mode | Output | Price | |
|---|---|---|---|
| Bundle (default) | Artifacts + manifest + hashes | $0.075 | |
| Lean | Text only | $0.010 | |
| Summarize | 3-sentence summary + key points | $0.018 | |
| Q&A | Answer + citations | $0.022 | |
| Claim-check | supported / contradicted / not_found + evidence | $0.025 | |
| Translate | Full translated document | $0.040 | |
| Structured | Extracted fields + citations | $0.025 | |
| Notarize | On-chain Base L2 tx hash | $0.050 | |
| Verify, download, history, signing keys, and all GET endpoints are free. | |||