Inference

cursorBench31

  • Not in index
  • Coding
ModelsModels with a result.7
Top resultBest result on this test.N/A
Top-3 spreadPoints from first to third.N/A
YearYear of release.N/A

Result and price

020406080100$1$3$10$30$100RESULTOUTPUT PRICE PER 1M TOKENS · LOG SCALE

Results

7 results
#ModelResultThe score from the source.
—Claude Fable 570.6%
—Composer 2.563.2%
—GPT-5.559.2%
—Claude Opus 4.858.4%
—Gemini 3.5 Flash49.8%
—Claude Sonnet 4.648.8%
—Kimi K2.647.6%