Text-to-Code Generation on PandasPlotBench
9.7Code Error RateJANUSCODER-14B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| JANUSCODER-14BModel Type=Open-Source, Parameters=14B2025.10 | 9.7 | 67 | |
| GPT-4oModel Type=Proprietary2025.10 | 9.7 | 72 | |
| Qwen2.5-Coder-32B-InsModel Type=Open-Source, Parameters=32B2025.10 | 12 | 66 | |
| Qwen3-14BModel Type=Open-Source, Parameters=14B2025.10 | 12.6 | 65 | |
| JANUSCODER-8BModel Type=Open-Source, Parameters=8B2025.10 | 14.9 | 63 | |
| Qwen3-8BModel Type=Open-Source, Parameters=8B2025.10 | 20 | 63 | |
| Qwen2.5-Coder-7B-InsModel Type=Open-Source, Parameters=7B2025.10 | 21.1 | 63 | |
| LLaMA3-8B-InstructModel Type=Open-Source, Parameters=8B2025.10 | 26.9 | 59 |