Skip to content

Instantly share code, notes, and snippets.

@encredible

encredible/LOGICKOR_REPORT.md

Created August 16, 2026 03:14
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save encredible/f6c6f47d0f2bf22a05a90fe8e0b4428f to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save encredible/f6c6f47d0f2bf22a05a90fe8e0b4428f to your computer and use it in GitHub Desktop.
Download ZIP
Gaiel Korean LLM Benchmark (LogicKor Results)
Raw
LOGICKOR_REPORT.md

Gaiel Series Korean Intelligence Benchmark (LogicKor)

Evaluated with GPT-4o Judge across 6 core Korean language & reasoning domains.

Overall Score Leaderboard

Model Single Turn Multi Turn Overall Score (10점 만점)
Gaiel-7B-Korean-Tuned-MLX 4.78 3.59 4.19
Gaiel-32B-Korean-Tuned-MLX 8.67 7.41 8.04

Category Breakdown

encredible/Gaiel-7B-Korean-Tuned-MLX

Category Single Turn Multi Turn
추론(Reasoning) 5.95 3.62
수학(Math) 5.81 4.43
글쓰기(Writing) 3.14 2.57
코딩(Coding) 4.47 3.26
이해(Understanding) 6.05 4.86
문법(Grammar) 3.24 2.76
Overall 4.78 3.59 (Avg: 4.19)

encredible/Gaiel-32B-Korean-Tuned-MLX

Category Single Turn Multi Turn
수학(Math) 9.57 9.05
추론(Reasoning) 8.33 5.57
글쓰기(Writing) 8.95 7.14
코딩(Coding) 9.43 8.62
이해(Understanding) 8.57 8.24
문법(Grammar) 7.14 5.86
Overall 8.67 7.41 (Avg: 8.04)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment