Benchmark
Capture-the-Flag Challenges (Internal) Benchmark
What is Capture-the-Flag Challenges (Internal)?
Capture-the-Flag Challenges is OpenAI's internal expansion of competitive, professional-level cybersecurity capture-the-flag tasks used to evaluate…
- Modality
- text
- Evaluates
- safety, agents, code
- Openness
- unknown
- Importer
- LLM Stats
- Reported scores
- 3
Source provenance
Capture-the-Flag Challenges (Internal) results and reported scores
LLM Stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| GPT-5.6 Luna | OpenAI | 0.852 | 2026-07-09 | evidence |
| GPT-5.6 Sol | OpenAI | 0.967 | 2026-07-09 | evidence |
| GPT-5.6 Terra | OpenAI | 0.918 | 2026-07-09 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.