New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
ARC-Easy benchmark
AI model leaderboard for the ARC-Easy benchmark. Compare how large language models score on ARC-Easy, see the full ranking, and understand what this AI benchmark measures. Marin 8B currently leads with 86.5. AI2 Reasoning Challenge (Easy set) — grade-school science questions answerable by simple retrieval.
Leaderboard
| # | Model | Organization | Score | Variant | Source |
|---|---|---|---|---|---|
| #1 | Marin 8B | Marin | 86.5 | base-from-hf-readme | official ↗ |
| #2 | Llama 3.1-8B | Meta AI | 85.8 | base-cited-marin | official ↗ |
| #3 | DCLM 7B | Apple | 82.2 | from-hf-readme | official ↗ |
Frequently asked questions about ARC-Easy
What is the ARC-Easy benchmark?
AI2 Reasoning Challenge (Easy set) — grade-school science questions answerable by simple retrieval.
How is the ARC-Easy benchmark scored?
ARC-Easy is scored using the accuracy metric, where a higher score is better. GenAIList aggregates reported scores from model providers and papers into a single ranked leaderboard.
Which AI model scores highest on ARC-Easy?
As of the latest reported scores on GenAIList, Marin 8B achieves the highest result on ARC-Easy with a score of 86.5.
Is a higher ARC-Easy score better?
Yes. On ARC-Easy a higher score indicates better performance, so models near the top of the leaderboard are the strongest.