zhipuch

Results 23 comments of zhipuch

> ![Image](https://github.com/user-attachments/assets/fadb7f59-d671-40fa-bffd-45a8cccaf866) > > The attached spreadsheet contains the results of my tests on [ScreenShotV2](https://hf-mirror.com/datasets/OS-Copilot/ScreenSpot-v2) , which include performance metrics for various categories such as mobile_text, mobile_icon, desktop_text, desktop_icon, web_text,...

你的数据中是包含纯文本还是说每条数据都有一张图片呢

> > 你的数据中是包含纯文本还是说每条数据都有一张图片呢 > > 每条数据都有一张图片。严格按照所提供的数据格式,形成的整个数据集。也就是和下面这个格式一模一样。两轮对话,第一个带图片,第二个不带。 { "messages": [ { "role": "user", "content": "图片中的动物是什么?", "image": "/root/images/0001.jpg" }, { "role": "assistant", "content": "图片中有一只猫。" }, { "role": "user", "content": "图片中的猫在做什么?" }, {...

glm4和glm3的分词方式、词表大小都不一样啊

也就是说,你使用的是chatglm3-6b,然后tokenizer部分改成glm4的方式了对吗?

这里确实是有点问题,已经提交pr了

启动openai_api_server服务,然后前端请求你的服务器地址不行吗

可以仔细看一下,有tool和没tool最终呢都只会有一个sys的