Inference
EN

cursorBench32

ModelsModels with a result.18
Top resultBest result on this benchmark.73.4%Claude Fable 5.1
Top-3 spreadPoints from first to third.2.9 pts
YearYear of release.—

Result and price

18 of 18 results

Results

18 results
#ModelResultThe score from the source.
01Claude Fable 5.173.4%
02Grok 4.670.8%
03Claude Fable 570.5%
04Claude Opus 570.0%
05Gemini 3.8 Flash69.2%
06GPT-5.6 Sol67.2%
07Grok 4.566.7%
08GPT-5.6 Terra64.9%
09Claude Opus 4.862.3%
10Claude Sonnet 561.5%
11GPT-5.6 Luna61.1%
12Kimi K360.8%
13GPT-5.558.4%
14Composer 2.556.1%
15GLM-5.255.0%
16Gemini 3.6 Flash53.5%
17Kimi K2.7 Code49.7%
18Gemini 3.5 Flash48.8%