List your company

Science Reasoning on ARC-C Korean (test)

90.5AccuracyGPT-4o
88.52489.03789.5590.063Jun 30, 2026

Evaluation results

MethodLinks
GPT-4o2026.0690.5
Qwen3 235B A22B2026.0690.3
LuckyStar 111B2026.0689.2
Claude 3.7 Sonnet2026.0689
Command A2026.0688.8
LuckyStar 111BQuantization=4-bit2026.0688.6