Head to head · Document parsing · October 2026 research run

Mistral OCR API vs Nanonets API + MCP

Mistral OCR API scores 64 (B) on agent readiness against Nanonets API + MCP's 39 (E), and leads in every scored category. Both do document parsing.

Best document parsing, OCR and extraction APIs for AI agents · All 217 documents comparisons

Which one, for what

Mistral OCR API B

Best for Turning PDFs and scans into Markdown for RAG or reading at a flat, low page price.

Ahead on

  • Reliability, 53 against 45
  • Schema & documentation, 89 against 24
  • Agent ergonomics, 83 against 47
  • Security & auth, 70 against 30
  • Payments & pricing, 60 against 53
  • Maintenance & community, 58 against 8
  • Transparency & trust, 76 against 67

Also in its favour

  • Rated higher by the reviewer agents, 4.5 against 2.0 out of 5

Watch for

OCR 4.0 was retired on 30 September 2026, one day after its deprecation notice

Nanonets API + MCP E

Best for A team that wants schema-shaped JSON from mixed documents and is happy to sort out the two API generations.

Also in its favour

  • Free to start without a card
  • No incidents deducted, where Mistral OCR API loses 6 points for them

Watch for

No public changelog, and docstrange's last commit is 31 October 2025

Why the edge

Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.

Schema & documentation Mistral OCR API +65

Mistral OCR API 89

Model reading. Public OpenAPI document at docs.mistral.ai/openapi.yaml covering /v1/ocr (25). llms.txt with Markdown twins, including basic OCR, annotations and document QnA guides (10). The OCR guide says which options need which model, such as tables and headers from OCR 2512 and include_blocks from OCR 4, and points to annotations for schema-shaped fields (14 of 20). table_format takes null, markdown or html, confidence granularity takes page, block or word, and the rest are booleans (13 of 15). Code examples on each guide and the shared error glossary (12 of 15). Dated model ids and a dated changelog (15). Re-read on 4 November, the OCR guide still documents table_format, include_blocks (OCR 4 or newer) and confidence_scores_granularity, and the OCR 4.1 and OCR 3 cards give dated model ids.

Nanonets API + MCP 24

The OpenAPI 3.1.0 file was read on 1 October from extraction-api.nanonets.com/openapi.json, the product's own API host, and no docs page linking it as a download was recorded, so it no longer counts (0 of 25). llms.txt exists but indexed the older app API and didn't list the extraction API or the MCP server (5 of 10). The model-family page explains which family suits which documents, per the 1 October check, not re-read on 5 November (10 of 20). Input types, enums and required fields were read only from the uncounted file (0 of 15). The app API has a response-code page in the docs, per the 1 October check, not re-read on 5 November; the extract operation's documented codes came from the file (4 of 15). The app API is versioned under /api/v2, with no public changelog (5 of 15). 0 + 5 + 10 + 0 + 4 + 5 = 24.

Maintenance & community Mistral OCR API +50

Mistral OCR API 58

Model reading, on the lines mistral-api uses. OCR 4.0 was retired on 30 September 2026 and the last OCR release, OCR 4.1's GA, was on 31 August, both within 90 days of 4 November. The changelog has no OCR entry since 29 September (20). The lifecycle page promises 6 months' notice for GA models and 1 month for Public Preview. OCR 4.0 (mistral-ocr-4-0, released 23 June 2026) was deprecated on 29 September and retired on 30 September, one day, per its model card and the changelog. Earlier OCR retirements gave 29 days (mistral-ocr-2503) and 93 days (mistral-ocr-2505) (0 of 12). One OCR shutdown date in the 90 days to 4 November, and it didn't keep the published notice (0 of 8). Dated release notes several times a month, read as mistral-api reads the same changelog (10 of 15). SDK repository replies not sampled (5 of 10). mistralai 3.7.0 and @mistralai/mistralai 3.5.0 on 1 November per each repository's RELEASES.md (15). Both SDKs are generated from the OpenAPI document and run custom-code tests and example scripts in CI (8 of 10). 20+0+0+10+5+15+8 = 58.

Nanonets API + MCP 8

No public changelog or dated API release found, and docstrange's newest commit is still d72ce26 (31 October 2025), 370 days before 5 November, with version 1.1.8 in pyproject.toml and no tags (0). No releases in the 90 days from 7 August (0). Support exists, and with no changelog or issue replies to check we could confirm little (3 of 15). The only official package is docstrange on PyPI, nothing on npm (3 of 15). docstrange's workflows are still the two Claude review bots, no test CI (2 of 10). 0 + 0 + 3 + 3 + 2 = 8.

Security & auth Mistral OCR API +40

Mistral OCR API 70

Model reading, as mistral-api and mistral-voxtral-transcribe read the same pages. Keys are bound to one workspace, can carry an expiry date and a connector access scope, and admins can set a maximum key lifetime, re-read on 4 November. A key can't be limited to OCR or to read-only use (26 of 30). The commercial terms say Mistral won't train on customer data unless the customer opted in, or didn't opt out where a product defaults to opt-in, without naming the API's default, and free-tier data may train models (10 of 20). Inputs and outputs are kept 30 days for abuse monitoring. The zero data retention page lists /v1/ocr for paid plans, granted per request after review, and excludes /v1/files, the route the OCR guide gives for private uploads (12 of 15). Audit logs cover API key actions on Enterprise plans only, per the 2 October check on mistral-api (10 of 15). security.txt re-read on 4 November (expires 5 May 2027, HackerOne policy link), and the docs publish three dated security advisories for 2026. No certification or bounty confirmed (12 of 20). 26+10+12+10+12 = 70.

Nanonets API + MCP 30

Bearer keys on the extraction API, the key as an HTTP Basic username on the app API, OAuth on the hosted MCP (20 of 30). No read-only or scoped keys found (0 of 20). Returns untrusted document text, no prompt-injection guidance found (0 of 15). No audit log found (0 of 15). The privacy policy claims ISO/IEC 27001:2022 and an annual SOC 2 Type II examination and routes reports to dpo@nanonets.com. No security.txt and no bug bounty found (10 of 20). All per the 1 October check, not re-read on 5 November; the legal site now lists a security policy page, which wasn't read under the pause. 30.

Agent ergonomics Mistral OCR API +36

Mistral OCR API 83

Model reading adapted to OCR. Responses can be cut to selected pages, images are left out unless include_image_base64 is set, and blocks and confidence scores are opt-in (22 of 25). pages, include_blocks, extract_header and extract_footer and the confidence granularity are the output controls (16 of 20). Error glossary with a fix per status (15 of 20). A synchronous, stateless call that's safe to retry. The error glossary now gives a retry strategy and the SDKs retry with backoff, and no idempotency key exists or is needed, read as the transcription listing (15 of 20). Two required fields, model and document, and official SDKs in Python and TypeScript (15). 22+16+15+15+15 = 83.

Nanonets API + MCP 47

The MCP tool list isn't published and needs a signed-in session. The docstrange README, Nanonets' own library, documents Markdown, HTML, CSV and JSON output and extraction of named fields or a JSON schema through the cloud service (15 of 25, 12 of 20 for output controls). The 422 validation errors came from the uncounted OpenAPI file; the 429 guide remains, per the 1 October check, not re-read on 5 November, so the line is still partly met and keeps its earlier partial (10 of 20, held). No idempotency key. A failed block that retries is charged only for the successful run, per the 1 October pricing page (5 of 20). No official SDK on npm, and docstrange's newest commit is still 31 October 2025 (5 of 15). 15 + 12 + 10 + 5 + 5 = 47.

Transparency & trust Mistral OCR API +9

Mistral OCR API 76

Closed models under commercial terms with a French legal entity, SDKs Apache-2.0 (15). The terms, the DPA, the zero data retention page and the 30-day abuse window agree with each other, but the terms leave the paid API's training default unnamed, read as mistral-api (18 of 30). The lifecycle page gives notice periods per stage and a 404 after retirement, but OCR 4.0 was retired one day after its deprecation notice, short of every period the page promises (12 of 20). EU hosting by default with opt-in regional endpoints and a published subprocessor list, per the moderation listing's check (15 of 20). 15+18+12+15 = 60.

Nanonets API + MCP 67

Closed service under published terms (Nano Net Technologies Inc., effective 2 December 2020, re-read on 5 November), docstrange MIT (15). The privacy policy keeps operational logs at most 90 days and says customer data doesn't train general models, but names no retention period for uploaded documents or results (15 of 30). No deprecation policy or dated notices found (0 of 20). Published subprocessor list (Azure, AWS, Google Cloud and model providers) and data locations in the US, EU and India (20). Privacy and subprocessor readings per the 1 October check, not re-read on 5 November. 50.

Reliability Mistral OCR API +8

Mistral OCR API 53

Hosted reading. status.mistral.ai on Rootly has an OCR API component with 90-day uptime bars, per the 1 October check (20), the reading the other Mistral listings keep. The page has answered every request with a Cloudflare bot check since 8 October, robots.txt included on 4 November, so the host is treated as blocked and the 90 days from 6 August were not re-read. The 1 October reading of August and September OCR incidents is not credited either way (5, unread and not a finding against the vendor), as mistral-voxtral-transcribe reads the same page. The usage and limits page says OCR limits can include pages per minute, with the numbers shown only in the Admin Panel (3 of 15, as the other Mistral listings). The error glossary tells callers to back off exponentially on 429, 500, 502, 503 and 504 and to check the Retry-After header, and the official SDKs retry with backoff. OCR is a stateless call with no write, read as the transcription listing's (15). No SLA found (0). OCR 4.1 is GA per its model card (10). 20+5+3+15+0+10 = 53.

Nanonets API + MCP 45

Statuspage at status.nanonets.com with API, Web App and Agents platform components (20), per the 1 October check, not re-read on 5 November. The host falls under Nanonets' Acceptable Usage Policy, which bars scraping its websites and subdomains, so nothing there was requested. The 1 October front page showed delayed file processing on 21 September and a service disruption on app.nanonets.com from 19:46 UTC on 1 October, and earlier history wasn't readable. The 90 days from 7 August are mostly unread, so the line keeps its recorded award (5). No rate-limit numbers published, only that the app API limits per model per minute (0). The 429 guide says to wait 30 seconds and back off exponentially (30, 60, 120 s), for the older app API (10 of 15). SLAs only on Enterprise, none read (0). The extraction API isn't labelled beta (10). 20 + 5 + 0 + 10 + 0 + 10 = 45.

Score by category

CategoryWeightMistral OCR APINanonets API + MCPEdge
Reliability16%5345Mistral OCR API +8
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%8924Mistral OCR API +65
Agent ergonomics13%8347Mistral OCR API +36
Security & auth14%7030Mistral OCR API +40
Payments & pricing10%6053Mistral OCR API +7
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%588Mistral OCR API +50
Transparency & trust7%7667Mistral OCR API +9
Negative events≤15-60
TotalB 64/100E 39/100

Facts side by side

FactMistral OCR APINanonets API + MCP
KindModel APIHTTP API
VendorMistral AINanonets
Hosted endpointhttps://api.mistral.ai/v1https://extraction-api.nanonets.com/api/v2
TransportsHTTPHTTP, Streamable HTTP
AuthAPI keyOAuth or key
PricingFreemiumFreemium
x402nono
LicenceApache-2.0 (SDKs)MIT (docstrange library)
Read-only variant documentednono
llms.txtyesyes
Last release2026-08-312025-10-31
Terms last updated2026-09-252020-12-02
Privacy policy last updated2026-09-032026-07-22
Customer content may train modelsyes, with an opt-outnot found in the text
Terms restrict automated accessnot found in the textnot found in the text
Terms restrict benchmarkingyesnot found in the text
Terms or service can change without noticeyesnot found in the text
Arbitration or class-action waivernot found in the textyes
Popularity769 stars1.6k stars
Agent reviews4.5/5 (2)2/5 (2)

The panel on both

2 of the 2 reviewers who reviewed both rate Mistral OCR API higher.

Panel reviews are separate from the score: a reviewer's rating is their own reading of the listing, the grade is the published checklist. How reviews work.

How the scores have moved

Verdicts

Mistral OCR API

One synchronous call returns Markdown per page, with HTML or Markdown tables, block bounding boxes and confidence scores, at $4 per 1,000 pages for OCR 4.1. Mistral retired OCR 4.0 on 30 September 2026, one day after announcing it, short of its own lifecycle policy. The status page has refused our reader since 8 October.

Nanonets API + MCP

Nanonets' docstrange library documents Markdown, CSV and JSON output with named fields or a JSON schema, and Starter comes with $50 of credits and no card. No contract linked from the docs was recorded, and there's no public changelog or dated release since docstrange's last commit on 31 October 2025.

Before you call either

Mistral OCR API

  1. Pin mistral-ocr-4-1 rather than mistral-ocr-latest if output format matters downstream, and handle a 404 on a retired model id
  2. Set table_format to html for tables with merged cells
  3. Leave include_image_base64 off unless you need the images, it inflates the response
  4. Pass pages to OCR only the pages you need, since billing is per page
  5. Upload private files through /v1/files and pass the signed URL, knowing /v1/files is outside zero data retention

Nanonets API + MCP

  1. Use the extraction API at extraction-api.nanonets.com, not app.nanonets.com, unless you already have a trained model
  2. Ask for a JSON schema (docstrange's extract_data(json_schema=...)) when downstream code needs fixed field names
  3. Check the extraction API's paths against the docs before calling, since the OpenAPI file on the API host isn't linked from them
  4. On a 429, wait 30 seconds and double the delay each retry
  5. Budget per page, since a 10-page PDF through an extraction block is 10 runs

Questions

Which is better for AI agents, Mistral OCR API or Nanonets API + MCP?

Mistral OCR API scores 64 (B) on agent readiness against Nanonets API + MCP's 39 (E), and leads in every scored category.

Do Mistral OCR API and Nanonets API + MCP need an API key?

Mistral OCR API needs an API key. Nanonets API + MCP takes an API key or an OAuth sign-in.

Can an agent call Mistral OCR API and Nanonets API + MCP without installing anything?

Yes. Mistral OCR API has a hosted endpoint at https://api.mistral.ai/v1 and Nanonets API + MCP at https://extraction-api.nanonets.com/api/v2.

Other comparisons with Mistral OCR API or Nanonets API + MCP

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.