Code Generation 벤치마크
Code Generation on Turbulence
CorrSc
- 2023-12-22 — GPT-4: CorrSc 0.848
| Rank | Model | CorrSc | Paper | Code | Year |
|---|---|---|---|---|---|
| 1 | GPT-4 | 0.848 | Turbulence: Systematically and Automatically Testing Instruction-Tuned Large Language Models for Code | shahinhonarvar/turbulence-benchmark | 2023 |
| 2 | GPT-3.5-Turbo | 0.617 | Turbulence: Systematically and Automatically Testing Instruction-Tuned Large Language Models for Code | shahinhonarvar/turbulence-benchmark | 2023 |
| 3 | CodeLlama:13B-4bit-quantised | 0.327 | Turbulence: Systematically and Automatically Testing Instruction-Tuned Large Language Models for Code | shahinhonarvar/turbulence-benchmark | 2023 |
| 4 | CodeLlama:7B-4bit-quantised | 0.289 | Turbulence: Systematically and Automatically Testing Instruction-Tuned Large Language Models for Code | shahinhonarvar/turbulence-benchmark | 2023 |
| 5 | Command | 0.063 | Turbulence: Systematically and Automatically Testing Instruction-Tuned Large Language Models for Code | shahinhonarvar/turbulence-benchmark | 2023 |