Google DeepMind

Gemini 3.6 Flash

• Framework: Frontier Safety Framework
• 1 safeguard
• 4 tests by the provider
• 2 source documents

Safeguards the provider describes

• Domain-specific safeguards (CBRN or cyber)

  To make Gemini 3.6 Flash more resistant to jailbreaks, we enhanced Frontier Safety safeguards in the domains of Chemical, Biological, Radiological, and Nuclear (CBRN) and cyber offense misuses.

  Google DeepMind · 21 July 2026 · Read the source

The provider's own tests

• image to text safety · The provider · Harmful content

  Image to Text Safety Automated content safety evaluation measuring safety policies 0% Lower is better

  Google DeepMind · 21 July 2026 · Read the source
• multilingual safety · The provider · Harmful content
  percentage point increase decrease vs gemini 3 5 flash: -5.45% change · Test design: lower numbers are the intended result

  Multilingual Safety Automated safety policy evaluation across multiple languages -5.45% Lower is better

  Google DeepMind · 21 July 2026 · Read the source
• text to text safety · The provider · Harmful content
  percentage point increase decrease vs gemini 3 5 flash: -1.35% change · Test design: lower numbers are the intended result

  Text to Text Safety Automated content safety evaluation measuring safety policies -1.35% Lower is better

  Google DeepMind · 21 July 2026 · Read the source
• unjustified-refusals · The provider · Misuse & jailbreak
  percentage point increase decrease vs gemini 3 5 flash: 0.25% change · Test design: lower numbers are the intended result

  Unjustified-refusals Automated evaluation measuring model’s ability to respond to borderline prompts while remaining safe +0.25% Lower is better

  Google DeepMind · 21 July 2026 · Read the source

No document we read states: a release date, whether a capability level was crossed, an independent or government test. No incident on record names this version.

Compare with other versions