reasoning benchmark
ARC-C Leaderboard
ARC-C is a multiple-choice question-answering benchmark containing grade-school level science questions that require advanced reasoning.
Updated Sep 7, 2026
Models34
Model coverage34
MetricScore
EvidenceB
ARC-C Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelXI | Score97.20% | Percentile100.00% | Participants34 | EvidenceC | Evaluated |
| Rank02 | ModelME | Score96.90% | Percentile96.97% | Participants34 | EvidenceC | Evaluated |
| Rank03 | ModelAN | Score96.40% | Percentile93.94% | Participants34 | EvidenceC | Evaluated |
| Rank04 | ModelME | Score94.80% | Percentile90.91% | Participants34 | EvidenceC | Evaluated |
| Rank05 | ModelAM Nova ProAmazon | Score94.80% | Percentile87.88% | Participants34 | EvidenceC | Evaluated |
| Rank06 | ModelAN | Score93.20% | Percentile84.85% | Participants34 | EvidenceC | Evaluated |
| Rank07 | ModelAL | Score93.00% | Percentile81.82% | Participants34 | EvidenceC | Evaluated |
| Rank08 | ModelAMNova LiteAmazon | Score92.40% | Percentile78.79% | Participants34 | EvidenceC | Evaluated |
| Rank09 | ModelMA | Score91.29% | Percentile75.76% | Participants34 | EvidenceC | Evaluated |
| Rank10 | ModelMI | Score91.00% | Percentile72.73% | Participants34 | EvidenceC | Evaluated |
| Rank11 | ModelAMNova MicroAmazon | Score90.20% | Percentile69.70% | Participants34 | EvidenceC | Evaluated |
| Rank12 | ModelAN | Score89.20% | Percentile66.67% | Participants34 | EvidenceC | Evaluated |
| Rank13 | ModelAL | Score85.70% | Percentile63.64% | Participants34 | EvidenceC | Evaluated |
| Rank14 | ModelMI | Score84.60% | Percentile60.61% | Participants34 | EvidenceC | Evaluated |
| Rank15 | ModelMI | Score83.70% | Percentile57.58% | Participants34 | EvidenceC | Evaluated |
| Rank16 | ModelME | Score83.40% | Percentile54.55% | Participants34 | EvidenceC | Evaluated |
| Rank17 | ModelME | Score78.60% | Percentile51.52% | Participants34 | EvidenceC | Evaluated |
| Rank18 | ModelMA | Score71.90% | Percentile48.48% | Participants34 | EvidenceC | Evaluated |
| Rank19 | ModelGO | Score71.40% | Percentile45.45% | Participants34 | EvidenceC | Evaluated |
| Rank20 | ModelCO | Score70.99% | Percentile42.42% | Participants34 | EvidenceC | Evaluated |
| Rank21 | ModelAC | Score70.50% | Percentile39.39% | Participants34 | EvidenceC | Evaluated |
| Rank22 | ModelAC | Score70.40% | Percentile36.36% | Participants34 | EvidenceC | Evaluated |
| Rank23 | ModelNV | Score69.20% | Percentile33.33% | Participants34 | EvidenceC | Evaluated |
| Rank24 | ModelAC | Score68.90% | Percentile30.30% | Participants34 | EvidenceC | Evaluated |
| Rank25 | ModelGO | Score68.40% | Percentile27.27% | Participants34 | EvidenceC | Evaluated |
| Rank26 | ModelAC | Score67.30% | Percentile24.24% | Participants34 | EvidenceC | Evaluated |
| Rank27 | ModelNR | Score65.53% | Percentile21.21% | Participants34 | EvidenceC | Evaluated |
| Rank28 | ModelGO | Score61.60% | Percentile18.18% | Participants34 | EvidenceC | Evaluated |
| Rank29 | ModelGO | Score61.60% | Percentile15.15% | Participants34 | EvidenceC | Evaluated |
| Rank30 | ModelAC | Score60.90% | Percentile12.12% | Participants34 | EvidenceC | Evaluated |
ARC-C Highlights
The leading models and scores on this benchmark.
ARC-C Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for ARC-C
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is ARC-C?
What ARC-C measures and how its scores work.
ARC-C is the Challenge Set of the AI2 Reasoning Challenge (ARC), comprising questions answered incorrectly by both retrieval-based and word co-occurrence algorithms.
It measures Score, reported as a ratio, on questions testing 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
- No
- Evaluation key
- overall
LLM Stats. 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 97.20%.
What are the top three models on ARC-C?
The current leaders are MiMo-V2.5-Pro (97.20%), Llama 3.1 405B Instruct (96.90%), and Claude 3 Opus (96.40%).
Which ARC-C model has the lowest official input price?
Nova Micro has the lowest matched official input price at $0.04 input / $0.14 output per 1M tokens.
Which models are fastest among ARC-C results?
The fastest matched records are Llama 3.1 8B Instruct (2,047.00 tok/s via Cerebras), Llama 3.1 70B Instruct (1,204.00 tok/s via Cerebras), and Llama 3.2 3B Instruct (171.50 tok/s via DeepInfra).
Does the highest score result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.