DeepSWE coding benchmark ranks GPT-5.5 first at 70% while DeepSeek-v4-pro trails at 8%

Positive users praise GPT-5.5 as nearly AGI and call Gemini's gains impressive, while negative users express sadness over DeepSeek-V4-Pro's low benchmark score and criticize its chip choices and tool-use limitations.

7 comments with sentiment. View comments.