每次把程序中断重启后都要去下一次模型

[longbai]

Compiling Operators(3/3) used: 2.34s eta: 0s Hello! How can I assist you today? I\'m sorry, could you please clarify what is your name I\'m sorry, could you please clarify your

hello

^C[e 0425 14:20:54.735719 00 log.cc:258] Caught SIGINT, quick exit (base) JittorLLMs on  main via 🐍 v3.10.9 took 52m30s ❯ (base) JittorLLMs on  main via 🐍 v3.10.9 ❯ python cli_demo.py chatglm [i 0425 14:20:58.884435 00 compiler.py:955] Jittor(1.3.7.13) src: /Users/long/miniconda3/lib/python3.10/site-packages/jittor [i 0425 14:20:58.896114 00 compiler.py:956] clang at /usr/bin/clang++(14.0.3) [i 0425 14:20:58.896153 00 compiler.py:957] cache_path: /Users/long/.cache/jittor/jt1.3.7/clang14.0.3/py3.10.9/macOS-13.3.1-axa8/AppleM1/default [i 0425 14:20:59.103762 00 init.py:227] Total mem: 16.00GB, using 5 procs for compiling. [i 0425 14:20:59.233455 00 jit_compiler.cc:28] Load cc_path: /usr/bin/clang++ [i 0425 14:20:59.382083 00 compile_extern.py:522] mpicc not found, distribution disabled. Downloading https://cg.cs.tsinghua.edu.cn/jittor/chat-glm/assets/build/checkpoints/pytorch_model-00007-of-00008.bin to /Users/long/.cache/jittor/jt1.3.7/clang14.0.3/py3.10.9/macOS-13.3.1-axa8/AppleM1/default/checkpoints/chat-glm/pytorch_model-00007-of-00008.bin 1%|█▊ | 11.7M/1.00G [00:16<27:23, 646kB/s]^C[e 0425 14:21:19.913774 00 log.cc:246] Caught SIGCHLD. Maybe out of memory, please reduce your worker size. si_errno: 0 si_code: 0 si_status: 0 , quick exit (base) JittorLLMs on  main via 🐍 v3.10.9 took 21s

[ZXiangQAQ]

使那个不是下载而是加载吧，用 cli 模式的话，程序前台运行，中断直接退出程序了，重新运行要加载模型，换 api 模式，程序后台运行，走 http 请求交互就行了