Atomic Network Attack Simulation (Atomic network attacks)
We show this table for reference; we do not rank on it.
Irregular's domain-level evaluation of network attack simulation capability.
Benchmark score on Atomic network attacks — November 13, 2026
We compile the Atomic network attacks rows from benchmark-owner or independent runs. GPT-6 Astra leads the table at 100%, followed by GPT-5.5 (100%) and GPT-5.6 Sol (98%). We do not use these results to rank models overall.
GPT-6 Astra
OpenAI
Run by Atomic network attacksGPT-5.5
OpenAI
Run by Atomic network attacksGPT-5.6 Sol
OpenAI
Run by Atomic network attacks3 modelsAgenticCurrentDisplay onlyUpdated November 13, 2026
Benchmark score table (3 models)
| Rank | Model / configuration | Score | Who ran it | Parameters (B) | Open / closed |
|---|---|---|---|---|---|
| 1 | GPT-6 AstraOpenAIRun by Atomic network attacks | 100% | Run by Atomic network attacks | Not reported | Closed |
| 2 | GPT-5.5OpenAIRun by Atomic network attacks | 100% | Run by Atomic network attacks | Not reported | Closed |
| 3 | GPT-5.6 SolOpenAIRun by Atomic network attacks | 98% | Run by Atomic network attacks | Not reported | Closed |
- Lab-reported
- Benchmark-run
- Independent
Among the reported Atomic network attacks rows, GPT-6 Astra is first at 100%. The third row is 2 points behind. The broader top-10 range is 2 points, so many of the published results sit in a relatively narrow band.
3 models have been evaluated on Atomic network attacks. The benchmark falls in the Agentic category. We keep these results separate from the weighted global scoring system, so they remain source-specific evidence. Atomic network attacks is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.
About Atomic network attacks
Year
2026
Tasks
Atomic cyber tasks
Format
Domain average
Difficulty
Network attack simulation
Atomic isolates cyber skill domains in controlled tasks. This network-attack-simulation average remains display only and is not combined with other domains.
Atomic network attacks is one of the cyber benchmarks we track. The cybersecurity benchmark comparison sets it beside the others, one table per security job.
Freshness and provenance
Version
Atomic network attacks 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 Atomic network attacks measure?
Irregular's domain-level evaluation of network attack simulation capability.
Which model scores highest on Atomic network attacks?
GPT-6 Astra by OpenAI currently leads with a score of 100% on Atomic network attacks.
How many models are evaluated on Atomic network attacks?
3 AI models have published results on Atomic network attacks in the BenchLM catalog.
Compare top models on Atomic network attacks
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 7,000+ subscribers.
One email each week. Unsubscribe anytime.