Gemini 3.7 Flash
26 values about Gemini 3.7 Flash from 2 documents, in 6 metric families. In each family, values from its own documents come first.
- Developer
- Google DeepMind
- Release date
- 13 Aug 2026The date of its first card in the dataset, the Gemini 3.7 Flash Model Card.
- Availability
- Public
- Its own documents
- Gemini 3.7 Flash Model Card · 13 Aug 2026
- Gemini 3.7 Flash Frontier Safety Framework Report · 13 Aug 2026
- Also reported in
- No other document reports a value about it
Data as of 26 Sep 2026 · Dataset v0.1 · Methodology v0.1 · Changelog
Values by metric family
Each value as the document printed it. Select a value for its source, its checks and its history.
KeyVerifiedUnverifiedDisputedCorrected read off a figure or stated in wordsA value opens its source and history.
M1 · Evaluation awareness 2 values
| Evaluation and metric | Value | Document |
|---|---|---|
| From its own documentsGemini 3.7 Flash Frontier Safety Framework Report · 13 Aug 2026 | ||
| Evaluation awarenessQualitative observation | Gemini 3.7 Flash Frontier Safety Framework Report13 Aug 2026 | |
| Situational awareness challengesChallenges solved | Gemini 3.7 Flash Frontier Safety Framework Report13 Aug 2026 | |
M3 · Sabotage and sandbagging 2 values
| Evaluation and metric | Value | Document |
|---|---|---|
| From its own documentsGemini 3.7 Flash Frontier Safety Framework Report · 13 Aug 2026 | ||
| Sandbagging checksEvidence of deliberate underperformance | Gemini 3.7 Flash Frontier Safety Framework Report13 Aug 2026 | |
| Stealth challengesChallenges solved | Gemini 3.7 Flash Frontier Safety Framework Report13 Aug 2026 | |
M4 · Misalignment audits 1 value
| Evaluation and metric | Value | Document |
|---|---|---|
| From its own documentsGemini 3.7 Flash Frontier Safety Framework Report · 13 Aug 2026 | ||
| Frontier Safety Framework determination (misalignment)Stealth and situational awareness TCL | Gemini 3.7 Flash Frontier Safety Framework Report13 Aug 2026 | |
M7 · Harmful compliance and over-refusal 6 values
| Evaluation and metric | Value | Document |
|---|---|---|
| From its own documents | ||
| Bio/chem weapons query violationsViolation rateCondition: with safeguardsGemini 3.7 Flash Frontier Safety Framework Report13 Aug 2026 | Gemini 3.7 Flash Frontier Safety Framework Report13 Aug 2026 | |
| Image to Text SafetyImage-to-text policy-violation rate deltaCondition: vs Gemini 3.6 FlashGemini 3.7 Flash Model Card13 Aug 2026 | Gemini 3.7 Flash Model Card13 Aug 2026 | |
| Multilingual SafetyMultilingual policy-violation rate deltaCondition: vs Gemini 3.6 FlashGemini 3.7 Flash Model Card13 Aug 2026 | Gemini 3.7 Flash Model Card13 Aug 2026 | |
| Text to Text SafetyPolicy-violation rate deltaCondition: vs Gemini 3.6 FlashGemini 3.7 Flash Model Card13 Aug 2026 | Gemini 3.7 Flash Model Card13 Aug 2026 | |
| ToneObjective tone of refusals deltaCondition: vs Gemini 3.6 FlashGemini 3.7 Flash Model Card13 Aug 2026 | Gemini 3.7 Flash Model Card13 Aug 2026 | |
| Unjustified refusalsUnjustified refusals deltaCondition: vs Gemini 3.6 FlashGemini 3.7 Flash Model Card13 Aug 2026 | Gemini 3.7 Flash Model Card13 Aug 2026 | |
M8 · Jailbreak robustness 1 value
| Evaluation and metric | Value | Document |
|---|---|---|
| From its own documentsGemini 3.7 Flash Frontier Safety Framework Report · 13 Aug 2026 | ||
| CBRN jailbreak template successObjective success rateCondition: with CBRN safeguards | Gemini 3.7 Flash Frontier Safety Framework Report13 Aug 2026 | |
M10 · Dangerous capabilities and risk determinations 14 values
| Evaluation and metric | Value | Document |
|---|---|---|
| From its own documentsGemini 3.7 Flash Frontier Safety Framework Report · 13 Aug 2026 | ||
| Bio bottleneck sub-stages red-teamingBio sub-stages scoring below 80 | Gemini 3.7 Flash Frontier Safety Framework Report13 Aug 2026 | |
| CBRN red-team upliftBIO-1 score gain over web baselineCondition: expert red team | Gemini 3.7 Flash Frontier Safety Framework Report13 Aug 2026 | |
| CBRN red-team upliftBIO-2 score gain over web baselineCondition: expert red team | Gemini 3.7 Flash Frontier Safety Framework Report13 Aug 2026 | |
| CBRN red-team upliftChem scenario 1 score gain over web baselineCondition: expert red team | Gemini 3.7 Flash Frontier Safety Framework Report13 Aug 2026 | |
| CBRN red-team upliftChem scenario 2 score gain over web baselineCondition: expert red team | Gemini 3.7 Flash Frontier Safety Framework Report13 Aug 2026 | |
| Frontier Safety Framework determinationCBRN TCL | Gemini 3.7 Flash Frontier Safety Framework Report13 Aug 2026 | |
| Frontier Safety Framework determinationCBRN uplift CCL | Gemini 3.7 Flash Frontier Safety Framework Report13 Aug 2026 | |
| Frontier Safety Framework determinationCyber CCL | Gemini 3.7 Flash Frontier Safety Framework Report13 Aug 2026 | |
| Frontier Safety Framework determinationHarmful manipulation CCL | Gemini 3.7 Flash Frontier Safety Framework Report13 Aug 2026 | |
| Frontier Safety Framework determinationML R&D CCL | Gemini 3.7 Flash Frontier Safety Framework Report13 Aug 2026 | |
| GRB internal research-engineering benchmarkAverage pass@1 | Gemini 3.7 Flash Frontier Safety Framework Report13 Aug 2026 | |
| Harmful manipulation efficacy studyParticipants enrolled | Gemini 3.7 Flash Frontier Safety Framework Report13 Aug 2026 | |
| Manipulative cue rateTurns with manipulative cues | Gemini 3.7 Flash Frontier Safety Framework Report13 Aug 2026 | |
| Manipulative cue rateTurns with manipulative cues (higher-stakes scenarios)Condition: higher-stakes scenarios | Gemini 3.7 Flash Frontier Safety Framework Report13 Aug 2026 | |
Risk determinations
The developer's formal decisions about Gemini 3.7 Flash under its framework, as printed. Levels from different frameworks do not map onto one another.
| Domain | Level as printed | Framework and document |
|---|---|---|
| Bio/chem | alert_threshold_reached code, not the printed wording Frontier Safety Framework, version 3.1 Gemini 3.7 Flash Frontier Safety Framework Report · 13 Aug 2026 | Frontier Safety Framework, version 3.1 Gemini 3.7 Flash Frontier Safety Framework Report · 13 Aug 2026 |
| Bio/chem | not_reached code, not the printed wording Frontier Safety Framework, version 3.1 Gemini 3.7 Flash Frontier Safety Framework Report · 13 Aug 2026 | Frontier Safety Framework, version 3.1 Gemini 3.7 Flash Frontier Safety Framework Report · 13 Aug 2026 |
| Cyber | alert_threshold_reached code, not the printed wording Frontier Safety Framework, version 3.1 Gemini 3.7 Flash Frontier Safety Framework Report · 13 Aug 2026 | Frontier Safety Framework, version 3.1 Gemini 3.7 Flash Frontier Safety Framework Report · 13 Aug 2026 |
| AI R&D / autonomy | not_reached code, not the printed wording Frontier Safety Framework, version 3.1 Gemini 3.7 Flash Frontier Safety Framework Report · 13 Aug 2026 | Frontier Safety Framework, version 3.1 Gemini 3.7 Flash Frontier Safety Framework Report · 13 Aug 2026 |
| Manipulation | not_reached code, not the printed wording Frontier Safety Framework, version 3.1 Gemini 3.7 Flash Frontier Safety Framework Report · 13 Aug 2026 | Frontier Safety Framework, version 3.1 Gemini 3.7 Flash Frontier Safety Framework Report · 13 Aug 2026 |
| Misalignment | not_reached code, not the printed wording Frontier Safety Framework, version 3.1 Gemini 3.7 Flash Frontier Safety Framework Report · 13 Aug 2026 | Frontier Safety Framework, version 3.1 Gemini 3.7 Flash Frontier Safety Framework Report · 13 Aug 2026 |
“Code, not the printed wording”: the version 0 file stored a code for this level rather than the words the document printed. The wording will be read again from the source.