Benchmark
ARC-E Benchmark
What is ARC-E?
ARC-E (AI2 Reasoning Challenge - Easy Set) is a subset of grade-school level, multiple-choice science questions that requires knowledge and reasoning…
- Released
- 2018-03-14
- Modality
- text
- Evaluates
- reasoning, general
- Openness
- unknown
- Importer
- LLM Stats
- Reported scores
- 8
Source provenance
- Original evidence https://api.zeroeval.com/leaderboard/benchmarks/arc-e?top_n=500
ARC-E results and reported scores
LLM Stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| ERNIE 4.5 | Baidu | 0.607 | 2025-06-25 | evidence |
| Gemma 2 27B | 0.886 | 2024-06-27 | evidence | |
| Gemma 2 9B | 0.88 | 2024-06-27 | evidence | |
| Gemma 3n E2B | 0.758 | 2025-06-26 | evidence | |
| Gemma 3n E2B Instructed LiteRT (Preview) | 0.758 | 2025-05-20 | evidence | |
| Gemma 3n E4B | 0.816 | 2025-06-26 | evidence | |
| Gemma 3n E4B Instructed LiteRT Preview | 0.816 | 2025-05-20 | evidence | |
| Hermes 3 70B | Nous Research | 0.8295 | 2024-08-15 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.