News
GLM-5.1: AI Model Codes 8 Hours Straight (58.4 Score)
Z.ai's GLM-5.1 scores 58.4 on SWE-Bench Pro, beating GPT and Claude. Claims 8-hour autonomous coding. ...
