IFBench
Ai2's instruction-following benchmark that measures precise instruction-following generalization on 58 diverse, verifiable out-of-domain output constraints designed to test whether models can obey novel rules rather than overfit to familiar constraint templates.
What this benchmark measures
Ai2's instruction-following benchmark that measures precise instruction-following generalization on 58 diverse, verifiable out-of-domain output constraints designed to test whether models can obey novel rules rather than overfit to familiar constraint templates.
Rows on this page are sourced from public benchmark artifacts, leaderboard exports, or source-linked model reports. Each row keeps benchmark version, source model name, and available run details attached to the score.
The metric shown here is accuracy. It should be interpreted within IFBench, not compared as part of a site-wide ranking.
Frequently asked
What is IFBench?
Ai2's instruction-following benchmark that measures precise instruction-following generalization on 58 diverse, verifiable out-of-domain output constraints designed to test whether models can obey novel rules rather than overfit to familiar constraint templates. It is a reasoning benchmark measured by accuracy.
What does accuracy mean on IFBench?
IFBench reports accuracy (%); higher is better. Scores are shown only within IFBench and are never averaged with other benchmarks.
What is the top reported IFBench score?
Grok 4.3 has the top reported score on IFBench: 83.3% (accuracy).
Why do IFBench scores differ across runs?
Harness, scaffold, reasoning effort, and prompt setup change results, so two runs of the same model can differ. evals.report keeps each score with its run context so the differences stay visible.
Does evals.report rank models across benchmarks?
No. IFBench scores are shown within their own metric; evals.report never combines benchmarks into a composite ranking or a single "best model".