CSAT 2027

Citation frequency 1/3Korean only
What does it measure?
A fun benchmark that has AI models sit the 2027 Korean College Scholastic Ability Test (CSAT, Suneung, held November 2026). It shows what score a model would get on the same paper Korean high-school seniors take.
Who evaluates it, and how?
Subjects: Korean, math, English, Korean history and all 17 exploration subjects (9 social, 8 science). Questions are given as OCR text (math in LaTeX) plus cropped figures, one whole section per request, through each vendor's official API with default settings and no system prompt or tools. The score is out of 450: Korean and math = common part + average of the elective parts, English (100) and Korean history (50) as is, exploration = subject average x 2. Questions and answer keys are not shown here.
Reading the score
Scores are shown as published by the source. Higher is better. Measured by an individual developer, hehee9 (CSAT-benchmark, MIT). It is a score for fun and is not used in our use-case rankings. The English listening part was replaced with its script. Measurement is still ongoing after the exam, so more models are added over time.
Related benchmarks
CSAT 2026
Top score450
Models tested36
Scoring basisHigher is betterMax score 450
Last updated2026.11.20
Model release date
9 models without a release date are not plotted; they appear only in the table below.
  1. 1
    Google2026.11.18 · high
    450
  2. 1
    OpenAI2026.11.17 · high
    450
  3. 3
    Google2026.09.30 · high
    450
  4. 3
    OpenAI2026.11.17 · high
    450
  5. 5
    Anthropic2026.11.09 · high
    449
  6. 5
    Anthropic2026.09.22 · high
    449
  7. 7
    OpenAI2026.10.26 · high
    446
  8. 8
    Anthropic2026.09.28 · high
    442
  9. 9
    Google2026.10.28 · high
    438
  10. 10
    Moonshot AI2026.10.23 · high
    438

The "harness" is the agent program the AI used to carry out the task. The same model can score very differently depending on its harness and reasoning effort.