Gemini 2.5 Flash Model Card
A model card by Google DeepMind about Gemini 2.5 Flash, published 26 Sep 2025. We recorded 6 values from it.
- Developer
- Google DeepMind
- Type
- Model card
- Model covered
- Gemini 2.5 Flash
- Published
- 26 Sep 2025
- 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).
No changelog
How we know Header read by extraction pass
Checked Not yet checked
No copy of the 26 Sep 2025 version held Version of Dec 2025 (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.
26 Sep 2025
First published version
Dec 2025
Copy retrieved 26 Sep 2026; header says updated Dec 2025
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
5 values
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Gemini 2.5 Flash | Image to Text SafetyImage-to-text policy-violation rate delta | thinking; vs Gemini 2.0 Flash | Safety evaluations table | Unverified | |
| Gemini 2.5 Flash | Instruction FollowingSafe instruction following delta | thinking; vs Gemini 2.0 Flash | Safety evaluations table | Unverified | |
| Gemini 2.5 Flash | Multilingual SafetyMultilingual policy-violation rate delta | thinking; vs Gemini 2.0 Flash | Safety evaluations table | Unverified | |
| Gemini 2.5 Flash | Text to Text SafetyPolicy-violation rate delta | thinking; vs Gemini 2.0 Flash | Safety evaluations table | Unverified | |
| Gemini 2.5 Flash | ToneObjective tone of refusals delta | thinking; vs Gemini 2.0 Flash | Safety evaluations table | Unverified |
M10 Dangerous capabilities and risk determinations
1 value
| Model | Evaluation | Condition | Value | Location | Checked |
|---|---|---|---|---|---|
| Gemini 2.5 Flash | Frontier Safety Framework determinationAll CCLs (inherited) | None stated | Frontier Safety | Unverified |
Extraction coverage
What we read of this document, and where each value was read.
Our note v0 card_date "2025-12" is the updated-header date; use 2025-09-26 as published date and record a Dec 2025 version
All 6 values were read in the document itself.
All 6 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).