Benchmark
Robust IF
Robust IF evaluates instruction-following robustness on diverse, hard prompts, measuring whether a model reliably adheres to constraints across…
- Modality
- text
- Categories
- reasoning, instruction_following
- Openness
- unknown
- Source
- llm_stats
- Reported scores
- 1
Reported scores
llm_stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| MAI-Code-1-Flash | Microsoft | 0.612 | 2026-06-02 | 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.