Code Summarization on Python Code Summarization downstream dataset
57.1CompScoreGPT-5-mini
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-5-miniInput Modality=Text2026.02 | 57.1 | |
| Gemini-3-ProInput Modality=Image2026.02 | 56.8 | |
| Qwen-3-VLInput Modality=Text2026.02 | 56.6 | |
| GPT-5.1Input Modality=Text2026.02 | 56.6 | |
| GPT-5-miniInput Modality=Image2026.02 | 56.5 | |
| Qwen-3-VLInput Modality=Image2026.02 | 56.4 | |
| Gemini-3-ProInput Modality=Text2026.02 | 56 | |
| GPT-5.1Input Modality=Image2026.02 | 55.9 | |
| Gemini-3-FlashInput Modality=Image2026.02 | 55.5 | |
| GLM-4.6vInput Modality=Text2026.02 | 55.4 | |
| Gemini-2.5-ProInput Modality=Image2026.02 | 55.2 | |
| Gemini-3-FlashInput Modality=Text2026.02 | 55.2 | |
| GLM-4.6vInput Modality=Image2026.02 | 54.6 | |
| Gemini-2.5-ProInput Modality=Text2026.02 | 54.5 |