Code Generation On Bigcodebench Instruct
Metriken
Pass@1
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Modellname | Pass@1 | Paper Title | Repository |
---|---|---|---|
DeepSeek-Coder-V2-Instruct | 48.2 | BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions | - |
GPT-4o-2024-05-13 | 51.1 | BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions | - |
0 of 2 row(s) selected.