other people's gguf file can not run

[xuedaobian]

Issue description

error loading model: unknown model architecture: ''

Expected Behavior

While I was browsing Reddit, I found a GGUF model of phi-2 trained by others，but when i download the model from the repository ，it can not run 。

the reddit page: https://www.reddit.com/r/LocalLLaMA/comments/18hnhd6/tutorial_how_to_run_phi2_locally_or_on_colab_for/?rdt=58275

the resposity addr: https://huggingface.co/radames/phi-2-quantized/tree/main

when i execution npx --no node-llama-cpp chat --model ./models/model-v2-qk4.gguf and npx --no node-llama-cpp chat --model ./models/model-v2-q80.gguf, the error appear.

Actual Behavior

in this image, mistral-xxx is can run successful, it from thebloke

Steps to reproduce

i just run simple commands from getting started part, those included by npx --no node-llama-cpp chat --model and node index.js(code from getting stared part)

My Environment

Dependency	Version
Operating System
CPU	Apple M1
Node.js version	18.16.0
Typescript version	use javascript
node-llama-cpp version	2.8.2

Additional Context

i dont kow how to dstinguish which model can be run, its so hard

thank you receive

Relevant Features Used

• [ ] Metal support
• [ ] CUDA support
• [ ] Grammar

Are you willing to resolve this issue by submitting a Pull Request?

Yes, I have the time, and I know how to start.

[giladgd]

@xuedaobian There's something weird with this particular model that you tried, as I see there are multiple parts for each file, and I haven't seen GGUF model files like that before.

I think you better try to use this one instead: https://huggingface.co/kroonen/phi-2-GGUF/tree/main

[xuedaobian]

thank you @giladgd ,i visited the link of you sent, it's useful!