LogicKor - Single-Turn: leaderboard
LogicKor single-turn aggregate measuring Korean prompt performance without multi-turn context.
Metric: Score (0-10). Source: lk.instruct.kr. Saturation forecast: Estimated already saturated. 253 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o (2024-05-13) | 9.42 | #325 |
| 2 | Gemini 1.5 Pro (001) | 9.4 | #444 |
| 3 | GPT-4 Preview (0125) | 9.33 | #486 |
| 4 | GPT-4 Preview (1106) | 9.28 | #418 |
| 5 | Claude 3.5 Sonnet (20240620) | 9.26 | #299 |
| 6 | Gemini 1.5 Flash (001) | 9.02 | #577 |
| 7 | Claude 3 Opus (20240229) | 8.95 | #412 |
| 8 | Mistral Large 2 (Jul) | 8.9 | #484 |
| 9 | Qwen 2 72B Instruct | 8.76 | #576 |
| 10 | GPT-4 (0613) | 8.73 | #491 |
| 11 | GPT-4 Turbo | 8.66 | #404 |
| 12 | ko-gemma-2-9B-it | 8.59 | #541 |
| 13 | Claude 3 Sonnet (20240229) | 8.54 | #677 |
| 14 | Llama 3.1 405B Instruct FP8 | 8.54 | #389 |
| 15 | WizardLM-2 8x22B | 8.52 | #703 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=logickor-single-turn · How It Works · Data refreshed daily, snapshot 2026-10-20.