reasoning benchmark
AdvancedIF
AdvancedIF is a rubric-based benchmark measuring complex, multi-turn, and system-prompted instruction following ability, scored with a calibrated LLM judge against per-instruction rubrics.
Updated Aug 7, 2026
Published models2
Registry coverage2
MetricScore
EvidenceB
AdvancedIF leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | MI | 85.0% | 100.0% | 2 | C | |
| 2 | MI | 71.4% | 0.0% | 2 | C |
Score distribution
Top published rows on the benchmark's original scale.
AdvancedIF highlights
The top published results on this benchmark's own scale.
What is AdvancedIF?
Definition and scoring fields from the benchmark registry.
AdvancedIF is a rubric-based benchmark measuring complex, multi-turn, and system-prompted instruction following ability, scored with a calibrated LLM judge against per-instruction rubrics.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- AdvancedIF
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- advancedif|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about AdvancedIF.
Which model scores highest on AdvancedIF?
MAI-Thinking-1 is currently ranked first with 85.0%.
What does AdvancedIF measure?
AdvancedIF is a rubric-based benchmark measuring complex, multi-turn, and system-prompted instruction following ability, scored with a calibrated LLM judge against per-instruction rubrics.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
2 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.