Gemini 3 Pro Model Card

A model card by Google DeepMind about Gemini 3 Pro, published 18 Nov 2025. We recorded 7 values from it.

Developer: Google DeepMind
Type: Model card
Model covered: Gemini 3 Pro
Published: 18 Nov 2025
Original: The document at storage.googleapis.com (opens an external site)
Also at: Another address, deepmind.google (opens an external site)
Archived copy: No archived copy yet
Changelog: No changelog

Data as of 26 Sep 2026 · Dataset v0.1 · Methodology v0.1 · Changelog

Revisions

Each change we have recorded between two versions: the old and the new text, whether the document's changelog explains it, how we know, and whether we have checked it. A revision is not in itself evidence of wrongdoing; most are corrections (Methodology §7).

Marked updated; what changed is not recorded · 18 Nov 2025 to May 2026
Card marked "Last updated May 2026"
No changelog
How we know Header read by extraction pass
Checked Not yet checked
No copy of the 18 Nov 2025 version held Version of May 2026 (opens an external site)

Versions

Two versions are on record, oldest first. We hold a copy of one; the other is known only by its date.

18 Nov 2025
First published version
Date from

our source registry
Copy

Known to exist; no copy held
May 2026
Copy retrieved 26 Sep 2026; header says updated May 2026
Date from

the date in the document header
Copy

Copy retrieved on 26 Sep 2026
Address

This version at storage.googleapis.com (opens an external site)
Values

7 values recorded from this version
What changed 1 change recorded as a revision. See it in redline

No version has a file hash or an archived snapshot yet. From dataset v0.2 each retrieved version carries both (Methodology §7).

Values

Every value we recorded from this document, grouped by metric family and ordered by where the document prints it. Location is the section, table or page as the document numbers it. None has been blind-verified yet.

KeyVerifiedUnverifiedDisputedCorrected read off a figure or stated in wordsA value opens its source and history.

M7 Harmful compliance and over-refusal

6 values

M7 Harmful compliance and over-refusal: values in the Gemini 3 Pro Model Card
Model	Evaluation	Condition	Location	Checked
Gemini 3 Pro	Child safety launch thresholdsLaunch thresholds met	None stated	Safety evaluations	Unverified
Gemini 3 Pro	Image to Text SafetyImage-to-text policy-violation rate delta	vs Gemini 2.5 Pro	Safety evaluations table	Unverified
Gemini 3 Pro	Multilingual SafetyMultilingual policy-violation rate delta	vs Gemini 2.5 Pro	Safety evaluations table	Unverified
Gemini 3 Pro	Text to Text SafetyPolicy-violation rate delta	vs Gemini 2.5 Pro	Safety evaluations table	Unverified
Gemini 3 Pro	ToneObjective tone of refusals delta	vs Gemini 2.5 Pro	Safety evaluations table	Unverified
Gemini 3 Pro	Unjustified refusalsUnjustified refusals delta	vs Gemini 2.5 Pro	Safety evaluations table	Unverified

M8 Jailbreak robustness

1 value

M8 Jailbreak robustness: values in the Gemini 3 Pro Model Card
Model	Evaluation	Condition	Value	Location	Checked
Gemini 3 Pro	Jailbreak vulnerability assessmentStated risk vs prior model	None stated		Risks and mitigations	Unverified

Extraction coverage

What we read of this document, and where each value was read.

Our note We have not written a coverage note for this document yet.

All 7 values were read in the document itself.

Of the 7 values, 1 is a statement in words rather than a number; it is marked * and left out of charts by default.

All 7 values were extracted for version 0 of the dataset through a web reader, which did not always reach the later sections of long PDFs (Methodology §3).