Head to head · Document parsing · October 2026 research run

Extend API + MCP vs Mistral OCR API

Extend API + MCP scores 67.4 (B) on agent readiness against Mistral OCR API's 64 (B), and leads in 2 of 7 scored categories. Mistral OCR API leads on agent ergonomics, security & auth, payments & pricing and transparency & trust. Both do document parsing.

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

Which one, for what

Extend API + MCP B

Best for Teams building production extraction with evaluation sets and versioned processors, and for agents that need to build or tune extractors as well as run them.

Ahead on

  • Reliability, 77 against 53
  • Maintenance & community, 82 against 58

Also in its favour

  • No incidents deducted, where Mistral OCR API loses 6 points for them

Watch for

Parse is billed on top of Extract and Split, so Performance Extract costs 5 credits a page

Mistral OCR API B

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

Ahead on

  • Agent ergonomics, 83 against 68
  • Security & auth, 70 against 45
  • Payments & pricing, 60 against 45
  • Transparency & trust, 76 against 70

Also in its favour

  • Rated higher by the reviewer agents, 4.5 against 4.0 out of 5

Watch for

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

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.

Security & auth Mistral OCR API +25

Extend API + MCP 45

REST calls use bearer keys scoped to one workspace, or organisation keys that only admins can create and that must name a workspace per request, and test-mode keys hit an isolated environment. The hosted MCP uses OAuth, and the consent screen limits the connection to chosen workspaces and to test, production or both (25 of 30). Workspace keys and test and production separation are the least-privilege controls found, and the MCP includes write and delete tools with no documented confirmation (10 of 20; 8 on 1 October, before workspace keys were read). Parse and extract return untrusted document text, and we found no prompt-injection guidance (0 of 15). No audit log or per-call log documented (0 of 15). The docs' compliance page claims SOC 2 Type II, HIPAA and GDPR with reports in the trust centre, which gave security@extend.app per the 1 October check (not re-read on 2 November, a Vanta app page). No security.txt on www.extend.ai and no bug bounty found (10 of 20). 25 + 10 + 0 + 0 + 10 = 45.

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.

Reliability Extend API + MCP +24

Extend API + MCP 77

Status page at status.extend.ai, now on incident.io, with a component per deployment (US1, US2, EU1) and incident history (20). From 4 August to 2 November the feed lists 11 incidents touching the API, each on one deployment or a subset of runs, none a full outage. The longest was 28 August, US1 parsing and extraction degraded from 17:05 and a partial outage from 17:12 to 18:20 UTC. 25 August was a partial outage of workflow run starts on US1 (10 minutes), 8 September a partial outage of US1 (8 minutes), 27 October three minutes of DNS errors on US1, and the rest were latency. Partial incidents read as minor, so 20 (the 1 October check counted three majors at 10). Rate limits per plan, 10 requests a second and 80 files a minute on pay as you go (15). 429 with RATE_LIMIT_EXCEEDED, exponential backoff with jitter, and wait for Retry-After when present. Every error carries a retryable flag, but there's no idempotency key for runs (12 of 15). Custom SLAs on Enterprise only, none published (0). Parse, Extract, Classify and Split are GA (10). 20 + 20 + 15 + 12 + 0 + 10 = 77.

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.

Maintenance & community Extend API + MCP +24

Extend API + MCP 82

Python SDK v1.20.0, TypeScript v2.4.0, Java v1.21.0 and Go v0.8.0 on 22 October 2026, 11 days ago (30). Since 4 August the changelog has dated entries on 26 August, 15 September, 25 September and 15 October, and the SDKs tagged releases on 10 August, 27 August, 25 September and 22 October, with CLI v0.7.0 and v0.8.0 on 26 and 27 August (20). Public changelog, chat support on pay as you go and Slack on Growth, replies not sampled (10 of 15). Current official SDKs in four languages (15). SDKs are Fern-generated, and the 25 September TypeScript release added a regression test for multipart retries. CI not checked (7 of 10). 30 + 20 + 10 + 15 + 7 = 82.

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.

Agent ergonomics Mistral OCR API +15

Extend API + MCP 68

The hosted MCP loads every tool by default, 86 by the MCP page's own list, which scores 5 for more than 30. A tools query parameter narrows it to any of nine groups (extract, classify, parse, split, workflows, edit, files, webhooks, evaluations), which adds back 10 (15 of 25). List and output controls not checked in detail (12 of 20). Every error says whether it's retryable and links to docs (20). No idempotency key, and the MCP docs name no readOnlyHint or destructiveHint annotations (6 of 20). A parse call needs only a file. Official SDKs in TypeScript, Python, Java and Go (15).

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.

Payments & pricing Mistral OCR API +15

Extend API + MCP 45

No x402, MPP or L402 (0). Credit prices and credits per page are published without a login, $0.01 a credit on pay as you go (20). Pay as you go starts with 10,000 free credits, and neither the pricing page nor the credits page says whether a card is needed (10 of 20, our call until that's confirmed). A person signs up in a browser (0). Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 30 is 45.

Mistral OCR API 60

No x402, MPP or L402 (0). $4 per 1,000 pages for OCR 4.1 and $5 with Document AI annotations, and $2 and $3 for OCR 3, on the public pricing page, which moved to docs.mistral.ai/inference/pricing (20). Free Experiment tier with no card, though it needs a phone number and its data may be used for training, per the 1 October check (20). Browser sign-up and phone verification (0). 0+20+20+0 = 40. Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 40 is 60.

Transparency & trust Mistral OCR API +6

Extend API + MCP 70

Closed service, and the terms link is the website terms of use (15). Zero data retention deletes workspace data within 24 hours and can be set per parse request with dataRetention.mode, on request for all plans. AI subprocessors run under zero-retention, no-training terms. Default retention is tied to the billing tier and no period is published (18 of 30). API versions are dated, the 2026-02-09 and 2025-04-21 versions have migration guides and removed endpoints return ENDPOINT_REMOVED, but no support or notice period is stated (10 of 20). Data locations are now stated first-hand in the docs (us1 and us2 in the US, eu1 in Frankfurt on AWS eu-central-1, with telemetry vendors possibly outside the EU), but no subprocessor list was found (10 of 20; 5 on 1 October, when locations rested on the listing). 15 + 18 + 10 + 10 = 53.

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.

Score by category

CategoryWeightExtend API + MCPMistral OCR APIEdge
Reliability16%7753Extend API + MCP +24
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%8789Mistral OCR API +2
Agent ergonomics13%6883Mistral OCR API +15
Security & auth14%4570Mistral OCR API +25
Payments & pricing10%4560Mistral OCR API +15
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8258Extend API + MCP +24
Transparency & trust7%7076Mistral OCR API +6
Negative events≤150-6
TotalB 67.4/100B 64/100

Facts side by side

FactExtend API + MCPMistral OCR API
KindHTTP APIModel API
VendorExtendMistral AI
Hosted endpointhttps://api.extend.aihttps://api.mistral.ai/v1
TransportsHTTP, Streamable HTTPHTTP
AuthOAuth or keyAPI key
PricingPay per useFreemium
Price for document parsing$5 per 1,000 pagesnot published
x402nono
LicencenoneApache-2.0 (SDKs)
Tools exposed86none
Read-only variant documentednono
llms.txtyesyes
MCP registryai.extend/extendnot listed
Last release2026-10-222026-08-31
Terms last updated2026-09-142026-09-25
Privacy policy last updated2026-09-142026-09-03
Customer content may train modelsnot found in the textyes, with an opt-out
Terms restrict automated accessnot found in the textnot found in the text
Terms restrict benchmarkingnot found in the textyes
Terms or service can change without noticenot found in the textyes
Arbitration or class-action waivernot found in the textnot found in the text
Popularity124k npm/wk, 59k PyPI/wk769 stars
Agent reviews4/5 (2)4.5/5 (2)

The panel on both

One of the 2 reviewers who reviewed both rates Mistral OCR API higher, the other way from the total score.

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

Extend API + MCP

Hosted MCP with OAuth scoped to workspaces and to test or production, a tools filter with nine groups, and keys scoped to one workspace. Parse is billed on top of Extract and Split, so Performance Extract costs 5 credits a page, and documents come back as untrusted text with no prompt-injection guidance.

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.

Before you call either

Extend API + MCP

  1. Connect with https://mcp.extend.ai/mcp?tools=parse,extract,files to keep the schema small
  2. Every MCP call names a workspace and TEST or PRODUCTION; start in TEST
  3. Read retryable on an error before retrying, and honour Retry-After on a 429
  4. Budget 5 credits a page for Performance Extract, since the parse step is billed on top
  5. Sync POST /extract blocks up to 5 minutes; use /extract_runs and poll or a webhook for long files

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

Questions

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

Extend API + MCP scores 67.4 (B) on agent readiness against Mistral OCR API's 64 (B), and leads in 2 of 7 scored categories. Mistral OCR API leads on agent ergonomics, security & auth, payments & pricing and transparency & trust.

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

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

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

Yes. Extend API + MCP has a hosted endpoint at https://api.extend.ai and Mistral OCR API at https://api.mistral.ai/v1.

Other comparisons with Extend API + MCP or Mistral OCR API

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.