MiniCPM-V
MiniCPM-V copied to clipboard
OpenBMB
[BUG] <title> 按照官方教程编译的ollama，在run myMinicpmv2.6:latest时候，模型直接报错退出
是否已有关于该错误的issue或讨论？ | Is there an existing issue / discussion for this?
- [X] 我已经搜索过已有的issues和讨论 | I have searched the existing issues / discussions
该问题是否在FAQ中有解答？ | Is there an existing answer for this in FAQ?
- [X] 我已经搜索过FAQ | I have searched FAQ
当前行为 | Current Behavior
`(deploy) ycx@user-NF5468-M7-A0-R0-00:~/ycxGit/ollama-minicpm-v2.6/ollama$ ./ollama run myMinicpmv2.6:latest
Are U dying? Error: an unknown error was encountered while running the model `
期望行为 | Expected Behavior
No response
复现方法 | Steps To Reproduce
No response
运行环境 | Environment
- OS: ub 20.04
- Python: 3.10.14
- Transformers: 4.44.0
- PyTorch: 2.4.0
- CUDA (`python -c 'import torch; print(torch.version.cuda)'`): 12.1
(deploy) ycx@user-NF5468-M7-A0-R0-00:~/ycxGit/ollama-minicpm-v2.6/ollama$ ./ollama create myMinicpmv2.6 -f /data1/home/ycx/workspace/HF/minicpmv2_6.Modelfile
transferring model data
using existing layer sha256:74c42a7b0f4dec95ce1e06b0732ec555feefc2f1b09064622c4214de908ed9dd
using existing layer sha256:4f512bc0bcc4367e82f2b247dcef63691691a1df6be1315568d87dac2ae7859d
creating new layer sha256:cca9919e361fa4c9aad81c621bf95e3cd5350376292eaca21f3b6651eb910fb9
creating new layer sha256:04f33aecb9bbd85b5e4d20b569f4a761a4059d128d44f58cd0dc52e58b3e3787
creating new layer sha256:1adc2b47fbc5dcbcc575c2dd5c076b06417f9c9aafb09e402d3daf4795b883e8
writing manifest
success
备注 | Anything else?
No response
可以试试这个 https://ollama.com/library/minicpm-v:latest
I’m having no luck with this, I’m having this issue on ubuntu, and it happens for every model I attempt to load, here’s an example of me having this exact issue with phi3 as an example.
可以试试这个 https://ollama.com/library/minicpm-v:latest
I’m having no luck with this, I’m having this issue on ubuntu, and it happens for every model I attempt to load, here’s an example of me having this exact issue with phi3 as an example.
update ollama