Korean CSAT 2026 (Easy Mode) - Society and Culture: leaderboard

2026 Korean CSAT (college entrance exam, November 2025) solved by LLM configurations via official APIs without tools (hehee9/CSAT-benchmark), in Easy mode: one request per question, with OCR text, LaTeX and figures. Society and Culture inquiry subject (20 questions); points out of 50, higher is better.

Metric: Points (out of 50). Source: hehee9.github.io. Saturation forecast: Estimated already saturated. 151 models tracked.

Top models

#ModelScoreOverall rank
1Gemini 3.5 Flash (High)50#66 (Gemini 3.5 Flash)
2GPT-6 Astra (High)50#6 (GPT-6 Astra)
3Gemini 3 Flash (High)48#103 (Gemini 3 Flash)
4GPT-5.6 Sol (Max)47#22 (GPT-5.6 Sol)
5Gemini 3.1 Pro (Preview) (High)47#64 (Gemini 3.1 Pro (Preview))
6Gemini 3.7 Flash (High)47#24 (Gemini 3.7 Flash)
7Claude Opus 4.7 (High)47#51 (Claude Opus 4.7)
8Gemini 3.8 Flash (Low)47#30 (Gemini 3.8 Flash)
9Claude Opus 4.7 (Max)47#51 (Claude Opus 4.7)
10GPT-5.4 (High)46#83 (GPT-5.4)
11GPT-5.6 Terra (Max)46#40 (GPT-5.6 Terra)
12Grok 4.5 (High)46#54 (Grok 4.5)
13GPT-5.4 (xHigh)46#83 (GPT-5.4)
14GPT-5.1 Codex46#168
15Muse Spark 1.3 (High)46#31 (Muse Spark 1.3)

No result here: #3 Claude Opus 5.5, #7 GPT-6.1 Sol, #9 Gemini 4 Argon.

Interactive version: theaggregate.ai/benchmark?slug=korean-csat-2026-easy-mode-society-and-culture · How It Works · Data refreshed daily, snapshot 2026-11-05.