Benchmark
Internal Research Debugging Evaluation
What is Internal Research Debugging Evaluation?
The Internal Research Debugging Evaluation measures whether models can debug 41 real bugs from internal OpenAI research experiments (plus alignment-auditing…
- Modality
- text
- Evaluates
- reasoning, agents, code
- Openness
- unknown
- Importer
- LLM Stats
- Reported scores
- 3
Source provenance
Internal Research Debugging Evaluation results and reported scores
LLM Stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| GPT-5.6 Luna | OpenAI | 0.508 | 2026-07-09 | evidence |
| GPT-5.6 Sol | OpenAI | 0.683 | 2026-07-09 | evidence |
| GPT-5.6 Terra | OpenAI | 0.678 | 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.