Model

Codex

OpenAI · closed weights

openai.com

Coding

Coding
BenchmarkScoreMetricHarnessSourceConfidenceLink
PolyCodeEval20.7pass_rateN/AIndependentmediumarxiv.org

Multimodal

Multimodal
BenchmarkScoreMetricHarnessSourceConfidenceLink
Artisan Benchmark32.4collision_singleN/AIndependentmediumarxiv.org

Reproduced by the community

Loading community runs…