Artificial Analysis Cyber Index (AA Cyber Index)
We show this table for reference; we do not rank on it.
Artificial Analysis' composite index of cyber-defense capability: finding vulnerabilities in real codebases, reproducing and validating them, and patching them without breaking existing functionality. It averages CWE-Bench-AA, DeepsecBench-AA and CyberGym-E2E-AA.
Benchmark score on AA Cyber Index — October 29, 2026
We compile the AA Cyber Index rows from secondary reports. Kimi K3.1 Pro leads the table at 64.1%, followed by Grok 4.7 (56.4%) and MiMo-V2.6-Pro (56.1%). We do not use these results to rank models overall.
Kimi K3.1 Pro
Moonshot AI
Grok 4.7
xAI
MiMo-V2.6-Pro
Xiaomi
19 modelsAgenticCurrentDisplay onlyUpdated October 29, 2026
Benchmark score table (19 models)
| Rank | Model / configuration | Score | Parameters (B) | Open / closed |
|---|---|---|---|---|
| 1 | Kimi K3.1 ProMoonshot AI | 64.1% | Not reported | Pending |
| 2 | Grok 4.7xAI | 56.4% | Not reported | Closed |
| 3 | MiMo-V2.6-ProXiaomi | 56.1% | Not reported | Open |
| 4 | GPT-6 LunaOpenAI | 52.7% | Not reported | Closed |
| 5 | GLM-5.3-FlashZ.AI | 49.9% | Not reported | Open |
| 6 | Mistral Large 4Mistral | 49.5% | Not reported | Open |
| 7 | Step 5 PreviewStepFun | 47.7% | Not reported | Open |
| 8 | Kimi K3Moonshot AI | 41.3% | Not reported | Open |
| 9 | DeepSeek V4.1 Flash 1017DeepSeek | 40.8% | Not reported | Open |
| 10 | GPT-6 SolOpenAI | 36.8% | Not reported | Closed |
| 11 | GLM-5.3Z.AI | 36.4% | Not reported | Open |
| 12 | Beam 1.0 501B-A23BReflection AI | 33.7% | Not reported | Open |
| 13 | GPT-6 AstraOpenAI | 33.4% | Not reported | Closed |
| 14 | Claude Opus 5.5Anthropic | 28.7% | Not reported | Closed |
| 15 | Gemini 3.8 FlashGoogle | 25.1% | Not reported | Closed |
| 16 | Claude Fable 5.1Anthropic | 25.0% | Not reported | Closed |
| 17 | MiniMax M3MiniMax | 22.4% | Not reported | Open |
| 18 | Nemotron 3 UltraNVIDIA | 13.1% | Not reported | Open |
| 19 | Qwen3.8-27BAlibaba | 12.7% | Not reported | Open |
Among the reported AA Cyber Index rows, Kimi K3.1 Pro is first at 64.1%. The third row is 8.0 points behind. The broader top-10 range is 27.3 points, so the table still separates the published systems.
19 models have been evaluated on AA Cyber Index. The benchmark falls in the Agentic category. AA Cyber Index is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.
About AA Cyber Index
Year
2026
Tasks
Three defensive cybersecurity evaluations
Format
Index score (0-100)
Difficulty
Real-world vulnerability discovery and patching
Cyber Index v1, launched October 2026. Independently run by Artificial Analysis under its published harness. Display-only; not yet admitted as independent-run evidence.
Freshness and provenance
Version
AA Cyber Index 2026
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
Questions
What does AA Cyber Index measure?
Artificial Analysis' composite index of cyber-defense capability: finding vulnerabilities in real codebases, reproducing and validating them, and patching them without breaking existing functionality. It averages CWE-Bench-AA, DeepsecBench-AA and CyberGym-E2E-AA.
Which model scores highest on AA Cyber Index?
Kimi K3.1 Pro by Moonshot AI currently leads with a score of 64.1% on AA Cyber Index.
How many models are evaluated on AA Cyber Index?
19 AI models have published results on AA Cyber Index in the BenchLM catalog.
Compare top models on AA Cyber Index
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 6,100+ subscribers.
One email each week. Unsubscribe anytime.