| Benchmark | Model | Rank | Results |
|---|---|---|---|
| openapi-code-completion-on-openapi-code | Code Llama 7B, fine-tuned with document splitting | #1 | Correctness, avg., %: 34Correctness, max., %: 42… |
| openapi-code-completion-on-openapi-code | Code Llama 7B, fine-tuned at 4096 tokens | #2 | Correctness, avg., %: 32Correctness, max., %: 45… |
| openapi-code-completion-on-openapi-code | Code Llama 7B | #3 | Correctness, avg., %: 31.1Correctness, max., %: 36… |
| openapi-code-completion-on-openapi-code | GitHub Copilot | #4 | Correctness, avg., %: 29Correctness, max., %: 29… |