Get an audit

Head to head · Guard self host · October 2026 research run

Granite Guardian vs Presidio

Presidio scores 66 (B) on agent readiness against Granite Guardian's 60.1 (C), and leads in 3 of 7 scored categories. Granite Guardian leads on security & auth and transparency & trust. Both do guard self host.

Granite GuardianIBM · Model API C60.1/100#478 of 841 PresidioData Privacy Stack · SDK + MCP B66/100#281 of 841

Which one, for what

Granite Guardian C

Good for A team with a GPU that wants one English-language judge for harm, jailbreaks, RAG groundedness, function-call checks and house rules, under a permissive licence with no gate.

Ahead on

  • Security & auth, 58 against 53
  • Transparency & trust, 70 against 65

Watch for

Trained and tested on English only, per the model card

Presidio B

Good for Detecting and masking personal data in prompts, outputs, logs and images on the owner's own machines, with detection tuned by entity, threshold and custom recognisers.

Ahead on

  • Reliability, 78 against 56
  • Schema & documentation, 69 against 60
  • Maintenance & community, 63 against 47

Also in its favour

  • Open source

Watch for

The REST containers have no authentication by design. The FAQ says to put a gateway or proxy in front

Score by category

CategoryWeight this runGranite GuardianPresidioEdge
Reliability16%205678Presidio +22
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%16.26069Presidio +9
Agent ergonomics13%16.26969even
Security & auth14%17.55853Granite Guardian +5
Payments & pricing10%12.56060even
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8.84763Presidio +16
Transparency & trust7%8.87065Granite Guardian +5
Negative events≤1500
Total60.1 · C66 · B

Facts side by side

FactGranite GuardianPresidio
KindModel APISDK + MCP
VendorIBMData Privacy Stack
Hosted endpointno (local only)no (local only)
TransportsHTTPHTTP
AuthNoneNone
PricingFreeFree
x402nono
LicenceApache 2.0 for the weights and the repositoryMIT
Read-only variant documentednono
llms.txtnono
Last release2026-04-292026-07-22
Terms last updatedno document linkedno document linked
Privacy policy last updatedno document linkedno document linked
Customer content may train models
Terms restrict automated access
Terms restrict benchmarking
Terms or service can change without notice
Arbitration or class-action waiver
Popularity182 stars11k stars, 1.2M PyPI/wk

Verdicts

Granite Guardian

One ungated Apache 2.0 model judges harm, jailbreaks, RAG groundedness, function-call errors and custom criteria, with signed weights and published evaluation code. It is trained and tested on English only, each call checks one criterion, the 4.1 prompt format differs from 3.x, and IBM's watsonx.ai lists only the deprecated 3.0 model.

Presidio

MIT-licensed personal data detector with a public OpenAPI document, tests on Python 3.10 to 3.14 and about 1.2 million weekly PyPI downloads. The REST containers have no authentication, the project states no SLA or support, and it covers personal data only, with no prompt injection or content moderation checks.

Before you call either

Granite Guardian

  1. Append the <guardian> block as the final user message, with the mode line, ### Criteria: and ### Scoring Schema:. Copy the strings from the model card, because no package builds them
  2. Use the no-think instruction for gating and parse <score>. Think mode writes a reasoning trace first, and the card's examples allow up to 2,048 output tokens
  3. Treat yes as the criterion being met, which for built-in criteria means the risk is present. Treat a missing <score> tag as a failed check
  4. Pass retrieved text through documents= and tool schemas through available_tools= in apply_chat_template, not inside the message text
  5. Under Ollama, set num_ctx in the request options. IBM's docs say the default context is short and long requests are truncated

Presidio

  1. Install from PyPI or pull images from ghcr.io/data-privacy-stack. The mcr.microsoft.com/presidio-* images are no longer updated
  2. Download a spaCy model (python -m spacy download en_core_web_lg) before the first AnalyzerEngine() call, or use the Docker image
  3. Send both text and language to /analyze. A request missing either returns HTTP 500 with a JSON error field
  4. Pass entities and score_threshold to limit results. Many country-specific recognisers are disabled by default and need enabling in the registry YAML
  5. Keep the containers on a private network or behind your own authenticating proxy. They accept any caller

Questions

Which is better for AI agents, Granite Guardian or Presidio?

Presidio scores 66 (B) on agent readiness against Granite Guardian's 60.1 (C), and leads in 3 of 7 scored categories. Granite Guardian leads on security & auth and transparency & trust.

Can an agent call Granite Guardian and Presidio without installing anything?

No hosted endpoint is listed for Granite Guardian. No hosted endpoint is listed for Presidio.

Are Granite Guardian and Presidio open source?

No open-source release is listed for Granite Guardian. Presidio is open source (MIT).

Other comparisons with Granite Guardian or Presidio

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.

Get an auditSee a sample report