IFBench evaluates precise instruction-following generalization on 58 challenging, verifiable out-of-domain constraints. Unlike IFEval which tests familiar constraint types, IFBench specifically measures how well models follow novel instructions they haven't been optimized for, exposing overfitting to common instruction patterns.
Reference2025Active16 models
Data verified 27 Jul 2026 · Methodology 1.8.0
Benchmark score on Instruction Following Benchmark
IFBench evaluates precise instruction-following generalization on 58 challenging, verifiable out-of-domain constraints. Unlike IFEval which tests familiar constraint types, IFBench specifically measures how well models follow novel instructions they haven't been optimized for, exposing overfitting to common instruction patterns. Results stay tied to the exact model variant and evaluation system. Multiple systems for the same model use the best published score on this page; overall Lumina scoring uses the median of ranking-eligible rows.
What does Instruction Following Benchmark measure?
IFBench evaluates precise instruction-following generalization on 58 challenging, verifiable out-of-domain constraints. Unlike IFEval which tests familiar constraint types, IFBench specifically measures how well models follow novel instructions they haven't been optimized for, exposing overfitting to common instruction patterns.
Which model scores highest on Instruction Following Benchmark?
MAI-Thinking-1 by Microsoft currently leads with 85%.
How many models are evaluated on Instruction Following Benchmark?
16 models in the LuminaBench cohort have a qualifying score on this benchmark.
