reasoning benchmark
ARC-C
The AI2 Reasoning Challenge (ARC) Challenge Set is a multiple-choice question-answering benchmark containing grade-school level science questions that require advanced reasoning capabilities. ARC-C specifically contains questions that were answered incorrectly by both retrieval-based and word co-occurrence algorithms, making it a particularly challenging subset designed to test commonsense reasoning abilities in AI systems.
Updated Aug 11, 2026
Models34
Model coverage34
MetricScore
EvidenceB
ARC-C Ranking
Higher score ranks better on this benchmark.
ARC-C Score Distribution
A closer view of the leading scores on this benchmark.
ARC-C Highlights
The leading models and scores on this benchmark.
What is ARC-C?
What ARC-C measures and how its scores work.
The AI2 Reasoning Challenge (ARC) Challenge Set is a multiple-choice question-answering benchmark containing grade-school level science questions that require advanced reasoning capabilities. ARC-C specifically contains questions that were answered incorrectly by both retrieval-based and word co-occurrence algorithms, making it a particularly challenging subset designed to test commonsense reasoning abilities in AI systems.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- ARC-C
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- arc-c|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about ARC-C.
Which model scores highest on ARC-C?
MiMo-V2.5-Pro is currently ranked first with 97.2%.
What does ARC-C measure?
The AI2 Reasoning Challenge (ARC) Challenge Set is a multiple-choice question-answering benchmark containing grade-school level science questions that require advanced reasoning capabilities. ARC-C specifically contains questions that were answered incorrectly by both retrieval-based and word co-occurrence algorithms, making it a particularly challenging subset designed to test commonsense reasoning abilities in AI systems.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
34 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.