List your company

Science Question Answering on ARC Korean (test)

92AccuracyTrinity Large
22.3240.4158.576.59Feb 16, 2026

Evaluation results

MethodLinks
Trinity LargeArchitecture=MoE, Evaluation Mode=Zero-shot2026.0292
Qwen3Size=8B, Evaluation Mode=Zero-shot2026.0287
Qwen3Size=4B, Evaluation Mode=Zero-shot2026.0281
DatologySize=8B, Evaluation Mode=Zero-shot2026.0269
TrillionLabsSize=7B, Evaluation Mode=Zero-shot2026.0268
DatologySize=3B, Evaluation Mode=Zero-shot2026.0262
Granite-4.0Size=Micro, Evaluation Mode=Zero-shot2026.0261
LFM2.5Size=1.2B, Evaluation Mode=Zero-shot2026.0261
Llama-3.1Size=8B, Evaluation Mode=Zero-shot2026.0259
SmolLM3Size=3B, Evaluation Mode=Zero-shot2026.0257
Llama-3.2Size=3B, Evaluation Mode=Zero-shot2026.0247
Llama-3.2Size=1B, Evaluation Mode=Zero-shot2026.0225