[BUG] <title> bug when converting hf to gguf for minicpm-v-2.6

[chongkuiqi]

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

About llama.cpp inference for minicpm-v 2.6 tutorial, the hf model should be converted to gguf, and the --minicpmv_version in minicpmv2_6-convert-image-encoder-to-gguf.py should be set 3 for minicpm-v-2.6, right ?

But the tutorial have minicpmv_version default 2, which is for minicpm-v-2.5

I find the similar problem in issue

期望行为 | Expected Behavior

No response

复现方法 | Steps To Reproduce

No response

运行环境 | Environment

- OS:
- Python:
- Transformers:
- PyTorch:
- CUDA (`python -c 'import torch; print(torch.version.cuda)'`):

备注 | Anything else?

No response

[LDLINGLINGLING]

Hello, this tutorial is the version of Minicpmv2.6. You can convert it into gguf and reason by following the tutorial directly.

[cpp-qn]

I convert minicpm-v-2.6 on the latest minicpmv-main branch, just follow the instructions in README-minicpmv2.6.md