Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Instruction Following benchmark
IHEval: instruction hierarchy evaluation. Every model the catalog carries a published IHEval value for, ranked by that value.
Tests whether a model follows the priority order of system messages, user messages, conversation history and tool outputs, including when their instructions conflict. 2 models in the ModelScale catalog have a published value.
Each point is a model with a published IHEval value and complete token prices, on the 75% input / 25% output blend. Frontier models are spaced evenly along the cost axis.
2 models with a published IHEval value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Conflict-setting accuracy |
|---|---|---|---|
| 1 | NVIDIA | 84.79 | |
| 2 | NVIDIA | 82.40 |
Evidence key: Observed
Rows are ordered by the value IHEval: Evaluating Language Models on Following the Instruction Hierarchy published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.
What the benchmark measures and how its value is published, from the source’s own record.
IHEval (IHEval: instruction hierarchy evaluation) is an instruction following benchmark published by IHEval: Evaluating Language Models on Following the Instruction Hierarchy. Tests whether a model follows the priority order of system messages, user messages, conversation history and tool outputs, including when their instructions conflict. Its evaluation set is described by the source as aligned and conflicting instruction sets. The source rates its difficulty as instruction priority.
Each model's result is published by IHEval: Evaluating Language Models on Following the Instruction Hierarchy as conflict-setting accuracy. ModelScale prints that value to 2 decimal places, exactly as published, and never converts or rescales it. This page ranks the 2 models that carry a published value, highest value first; a model the source never scored is left out rather than ranked at zero. The source does not state whether a higher value is the better result, so neither does this page.
All leaderboards · Instruction Following capability leaderboard