Misc. bug: Some server response JSON still not restored

[CentricStorm]

Name and Version

version: 4291 (ce8784b)
built with MSVC 19.29.30157.0 for x64

Operating systems

Windows

Which llama.cpp modules do you know to be affected?

llama-server

Problem description & steps to reproduce

After #10722, there are still some issues when using the native /completion endpoint:

• The has_new_line property is set to 1 instead of true like it used to be.
• The model property is set to gpt-3.5-turbo-0613 even though this isn't the OpenAI-compatible API.

First Bad Commit

No response

Relevant log output

No response

[ggerganov]

PRs welcome

[MichelleTanPY]

Hi @ggerganov , I saw the 'good first issue' tag on this. I'm new to this project and thought to give this a try. I manage to build the llama-server, but I'm getting this strange error when get the server up and running.
I'm following the README here : https://github.com/ggerganov/llama.cpp/tree/master/examples/server and getting the following error.
llama_model_load: error loading model: missing tensor 'token_embd.weight' llama_load_model_from_file: failed to load model common_init_from_params: failed to load model '.\models\ggml-vocab-deepseek-llm.gguf' srv load_model: failed to load model, '.\models\ggml-vocab-deepseek-llm.gguf' main: exiting due to model loading error
Any pointers on how I should proceed? Building with MSVC 19.42.34435.0 x64 Windows machine.

[ngxson]

• The model property is set to gpt-3.5-turbo-0613 even though this isn't the OpenAI-compatible API.

This is to prepare for adding OAI-compat to /v1/completions endpoint. If you're searching for the model path, please have a look at /props endpoint