IHEval: instruction hierarchy evaluation (IHEval)
We show this table for reference; we do not rank on it.
Tests whether a model follows the priority order of system messages, user messages, conversation history and tool outputs, including when their instructions conflict.
Benchmark score on IHEval — October 29, 2026
We compile the IHEval rows from provider self-reports. The table contains 2 models. Nemotron 3.5 Bolt E4B BF16 has the highest published score at 84.79%, but the available coverage and evaluation setups do not establish a market leader. We do not use these results to rank models overall.
Nemotron 3.5 Bolt E4B BF16
NVIDIA
Nemotron 3.5 Bolt E4B
NVIDIA
2 modelsInstruction FollowingCurrentDisplay onlyUpdated October 29, 2026
Benchmark score table (2 models)
| Rank | Model / configuration | Score | Parameters (B) | Open / closed |
|---|---|---|---|---|
| 1 | Nemotron 3.5 Bolt E4B BF16NVIDIA | 84.79% | Not reported | Open |
| 2 | Nemotron 3.5 Bolt E4BNVIDIA | 82.40% | Not reported | Open |
How to read this leaderboard
Operator receipt: 2 sourced rows are currently displayable on this page; the highest published score among these 2 models is Nemotron 3.5 Bolt E4B BF16 at 84.79%, which does not establish a market leader.
Honest limit: Only NVIDIA-run results are stored so far, from one harness. Display-only and excluded from weighted ranking.
About IHEval
Year
2025
Tasks
Aligned and conflicting instruction sets
Format
Conflict-setting accuracy
Difficulty
Instruction priority
NVIDIA reports it for Nemotron 3.5 Bolt E4B (October 23, 2026), run in its NeMo Gym harness, as accuracy on the conflict setting in reasoning mode.
Freshness and provenance
Version
IHEval 2025
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
Questions
What does IHEval measure?
Tests whether a model follows the priority order of system messages, user messages, conversation history and tool outputs, including when their instructions conflict.
Which model scores highest on IHEval?
Nemotron 3.5 Bolt E4B BF16 by NVIDIA currently leads with a score of 84.79% on IHEval.
How many models are evaluated on IHEval?
2 AI models have published results on IHEval in the BenchLM catalog.
Compare top models on IHEval
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 6,100+ subscribers.
One email each week. Unsubscribe anytime.