Python Code Clone Detection downstream dataset
81.6AccuracyGLM-4.6v
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GLM-4.6vInput Modality=Text2026.02 | 81.6 | 78.4 | |
| GPT-5.1Input Modality=Image2026.02 | 71.6 | 62.4 | |
| Gemini-3-ProInput Modality=Text2026.02 | 71 | 60.8 | |
| Gemini-3-ProInput Modality=Image2026.02 | 70.2 | 58.8 | |
| Gemini-3-FlashInput Modality=Text2026.02 | 70 | 59.4 | |
| GLM-4.6vInput Modality=Image2026.02 | 69.6 | 58.2 | |
| Qwen-3-VLInput Modality=Text2026.02 | 67.8 | 52.2 | |
| Gemini-3-FlashInput Modality=Image2026.02 | 67.8 | 55.4 | |
| Qwen-3-VLInput Modality=Image2026.02 | 67.2 | 51.2 | |
| Gemini-2.5-ProInput Modality=Image2026.02 | 67 | 50.6 | |
| GPT-5.1Input Modality=Text2026.02 | 65.8 | 46.8 | |
| Gemini-2.5-ProInput Modality=Text2026.02 | 65.4 | 46.4 | |
| GPT-5-miniInput Modality=Image2026.02 | 64.8 | 47 | |
| GPT-5-miniInput Modality=Text2026.02 | 59.4 | 33.2 |