Benchmark Almanac

Real-world vulnerability discovery

Gemini 4 Argon 86% In use

Agents search popular open-source codebases to rediscover about 1,200 recent confirmed vulnerabilities across many languages. Reports share discovered.

In-house Google DeepMind evaluation. The tasks are not public.

Made by Google DeepMind Finding and exploiting real bugs Tasks kept private Last in a launch post Sep 2026 Last model added Oct 2026
Level
Expert Specialists with years of training, often over hours of work
Scored
Checked automatically A program checks the result against an answer key, unit tests or the expected final state.
Answers
Agent task Many steps in a terminal, browser, desktop, app, game or robot.
2
Models scored in lab cards
1
Launch post lists it, from 1 lab
1
System card reports it, from 1 lab

Gemini 4 Argon leads with 86%

Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.

  1. 86.0 Gemini 4 Argon
  2. 71.0 Gemini 3.8 Flash Cyber
  1. 1 Gemini 4 Argon 86%
  2. 2 Gemini 3.8 Flash Cyber 71%
#ModelScoreSettingDateSource
1 Gemini 4 Argon 86% Oct 2026 Model card
2 Gemini 3.8 Flash Cyber 71% Sep 2026 Model card

The best score passed 85% in Oct 2026

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Reported by the lab in its model card Best so far
0% 20% 40% 60% 80% 100% Near the ceiling (85%) Gemini 4 Argon Gemini 3.8 Flash Cyber 0% 20% 40% 60% 80% 100% Near the ceiling (85%) Gemini 4 Argon Gemini 3.8 Flash Cyber

One lab put it in a launch post in the last six months

Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.

Google DeepMind
DateLaunch postScoreAs the post writes it
2 Sep 2026 Gemini 3.8 Flash Cyber 71 Real-world vulnerability discovery, internal benchmark across 20 languages

One system card reports it

Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.

DateDocumentScoreAs the document writes itSection
13 Oct 2026 Gemini 4 Argon system card 86 Real-world vulnerability discovery (internal), historical vulnerabilities discovered 86% (7.1.2) Cyber risk

More in Cybersecurity

All 111 in Cybersecurity
  • ExploitGym In use 869 real-vulnerability exploit-development tasks (userspace, V8, Linux kernel). Success means capturing a flag through the intended vulnerability. GPT-6 Astra 42.4%
  • ExploitBench In use 41 recent V8 JavaScript-engine N-day vulnerabilities. Progress measured by 16 grader-verified capability flags across five tiers. GPT-6 Astra 100%
  • CWE-bench In use Agents audit a repository and write a patch for a security flaw, without being told where the bug is. Covers six programming languages. Gemini 4 Carbon 71%
  • Artificial Analysis Cyber Index In use Agents do enterprise cyber defense tasks from three benchmarks, CWE-Bench-AA, DeepsecBench-AA and CyberGym-E2E-AA, combined into one success rate. Mistral Large 4 (Le Chonk, open weights) 51%
  • Big Sleep Evaluation (Google) In use Agents search large real codebases such as Chrome and Safari for critical, hard to find vulnerabilities. Scored by pass@1 success rate, run without safety guardrails. Gemini 3.5 Flash Cyber 72%
  • Chrome Production Commit Scanning (Google) In use Models scan commits in Chrome's production pipeline for vulnerabilities that were never publicly disclosed. Scored by pass@1 success rate. Gemini 3.5 Flash Cyber 72%