Google DeepMind

Gemini 3.7 Flash

• Framework: Frontier Safety Framework (April 2026)
• 5 statements about capability levels
• 8 tests by the provider
• 2 source documents

What the provider states

• Autonomy & replication · CCL (Critical Capability Level) · Stated as not crossed

  the model does not reach the ccl alert threshold.

  Google DeepMind · 13 August 2026 · Read the source
• Chemical, biological, radiological and nuclear · CCL (Critical Capability Level) · Stated as not crossed

  we can rule out the ccl for the cbrn domain with reasonable confidence based on the results from our testing.

  Google DeepMind · 13 August 2026 · Read the source
• Chemical, biological, radiological and nuclear · TCL (Tracked Capability Level) · Stated as not crossed

  we can rule out the tcl for the cbrn domain with reasonable confidence based on the results from our testing.

  Google DeepMind · 13 August 2026 · Read the source
• Cyber · CCL (Critical Capability Level) · Stated as not crossed

  gemini 3.7 flash reaches the alert threshold for this ccl, but not the ccl.

  Google DeepMind · 13 August 2026 · Read the source
• Deception & sabotage · TCL (Tracked Capability Level) · Stated as not crossed

  the model does not reach the tcl.

  Google DeepMind · 13 August 2026 · Read the source

The provider's own tests

• automated safety evaluations · The provider · Misuse & jailbreak

  we expect variation in our automated safety evaluations results, which is why we review flagged content to check for egregious or dangerous material.

  Google DeepMind · 13 August 2026 · Read the source
• cbrn uplift level 1 ccl · The provider · Chemical, biological, radiological and nuclear

  expert red teaming demonstrated a modest capability uplift over web baselines and a subset of experts were able to elicit accurate and actionable information across the full harm journey for both tested scenarios, prompting us to assess that the model has reached the alert threshold for this ccl.

  Google DeepMind · 13 August 2026 · Read the source
• image to text safety · The provider · Harmful content

  image to text safety automated content safety evaluation measuring safety policies no change lower is better

  Google DeepMind · 13 August 2026 · Read the source
• multilingual safety · The provider · Harmful content

  multilingual safety automated safety policy evaluation across multiple languages -0.48pp lower is better

  Google DeepMind · 13 August 2026 · Read the source
• situational awareness · The provider · Autonomy & replication

  gemini 3.7 flash is observant enough to correctly assess when it is in a testing environment, but it cannot successfully bypass testing restrictions.

  Google DeepMind · 13 August 2026 · Read the source
• text to text safety · The provider · Harmful content

  text to text safety automated content safety evaluation measuring safety policies +1.17pp lower is better

  Google DeepMind · 13 August 2026 · Read the source
• tone 1 · The provider · Misuse & jailbreak

  tone 1 automated evaluation measuring objective tone of model refusal -0.47pp higher is better

  Google DeepMind · 13 August 2026 · Read the source
• unjustified-refusals · The provider · Misuse & jailbreak

  unjustified-refusals automated evaluation measuring model’s ability to respond to borderline prompts while remaining safe +0.84pp lower is better

  Google DeepMind · 13 August 2026 · Read the source

No document we read states: a release date, an independent or government test, a safeguard. No incident on record names this version.

Some documents we hold are stored in lower case only, so their quotes are shown in lower case. The source link opens the original.

Compare with other versions