simplew2011

a llm/vlm usage case

3

Does TorchView support analysis of large models, such as qwen2.5-7b

- https://qwen.readthedocs.io/en/latest/deployment/vllm.html#thinking-non-thinking-modes - how to set this params 

FRAMES/Codeforces/CNMO dataset eval

1

## 功能描述 / Feature Description support some dataset describe in deepseek-r1：https://github.com/deepseek-ai/DeepSeek-R1/blob/main/README.md?plain=1#L118 - FRAMES dataset - https://zhuanlan.zhihu.com/p/735394464 - Codeforces - CNMO 2024

## 功能描述 / Feature Description - FewCLUE/cluewsc数据集：http://opencompass.oss-cn-shanghai.aliyuncs.com/datasets/data/FewCLUE.zip"

new model support

4

- [ ] DeepseekV3 - [x] Qwen3-30B-A3B - [ ] Qwen3-Next-80B-A3B-Instruct - [x] Qwen3_VL

前端一直在转圈，无法输入

10

- 服务器命令行方式运行正常：`uv run main.py` - 服务器webui方式启动，`./bootstrap.sh -d` - 局域网其它机器，访问下面两个ip均无法正常输入，输入框在转圈，页面和后台也没报错 - 点击示例输入，没前端反应，后台已接收到消息 - 网页访问机器和服务器可ping通 ``` ▲ Next.js 15.3.0 (Turbopack) - Local: http://localhost:3000 - Network: http://192.168.30.127:3000 ✓ Starting... ✓ Ready in 5.9s...