"llama.cpp error: 'error loading model architecture: unknown model architecture: 'deepseek2'

[ZeroCool22]

Prerequisites

Please answer the following questions for yourself before submitting an issue.

• [x] I am running the latest code. Development is very rapid so there are no tagged versions as of now.
• [x] I carefully followed the README.md.
• [ ] I searched using keywords relevant to my issue to make sure that I am creating a new issue that is not already open (or closed).
• [x] I reviewed the Discussions, and have a new bug or useful enhancement to share.

{ "title": "Failed to load model", "cause": "llama.cpp error: 'error loading model architecture: unknown model architecture: 'deepseek2''", "errorData": { "n_ctx": 8192, "n_batch": 512, "n_gpu_layers": 31 }, "data": { "memory": { "ram_capacity": "31.91 GB", "ram_unused": "26.74 GB" }, "gpu": { "gpu_names": [ "NVIDIA GeForce GTX 1080 Ti" ], "vram_recommended_capacity": "11.00 GB", "vram_unused": "9.98 GB" }, "os": { "platform": "win32", "version": "10.0.19045", "supports_avx2": true }, "app": { "version": "0.2.24", "downloadsDir": "C:\Users\ZeroCool22\.cache\lm-studio\models" }, "model": {} } }```

[ohmerhe]

+1

[coded-aesthetics]

+1

[Sid-Turner-Ellis]

+1

[justDeek]

I fixed it by updating Ollama (-> start the Ollama executable, right-click the icon in the notification area and click on 'restart to update'; or just reinstall Ollama).

[ZeroCool22]

I don't use Ollama at all, don't need that if i use LMS.

[AaronCodesPython]

I have the same error using, LM Studio 0.2.27, so it seems that this is an LMS problem.

[LeadhuntAI]

same issue here.. sad, wanted to test it finally, hope we will get an update for that soon!

[holchan]

Same

llama_model_load: error loading model: error loading model architecture: unknown model architecture: 'deepseek2' llama_load_model_from_file: failed to load model Traceback (most recent call last): File "/opt/conda/lib/python3.10/runpy.py", line 196, in _run_module_as_main return _run_code(code, main_globals, None, File "/opt/conda/lib/python3.10/runpy.py", line 86, in _run_code exec(code, run_globals) File "/opt/conda/lib/python3.10/site-packages/llama_cpp/server/main.py", line 97, in main() File "/opt/conda/lib/python3.10/site-packages/llama_cpp/server/main.py", line 83, in main app = create_app( File "/opt/conda/lib/python3.10/site-packages/llama_cpp/server/app.py", line 147, in create_app set_llama_proxy(model_settings=model_settings) File "/opt/conda/lib/python3.10/site-packages/llama_cpp/server/app.py", line 68, in set_llama_proxy _llama_proxy = LlamaProxy(models=model_settings) File "/opt/conda/lib/python3.10/site-packages/llama_cpp/server/model.py", line 31, in init self._current_model = self.load_llama_from_model_settings( File "/opt/conda/lib/python3.10/site-packages/llama_cpp/server/model.py", line 217, in load_llama_from_model_settings _model = create_fn( File "/opt/conda/lib/python3.10/site-packages/llama_cpp/llama.py", line 338, in init self._model = _LlamaModel( File "/opt/conda/lib/python3.10/site-packages/llama_cpp/_internals.py", line 57, in init raise ValueError(f"Failed to load model from file: {path_model}") ValueError: Failed to load model from file: /kaggle/working/QA-Pilot/llamacpp_models/DeepSeek-Coder-V2-Lite-Instruct-Q8_0_L.gguf

[justchillbruh]

For anyone still experiencing this issue with LM Studio, I've been able to use the DeepSeek Coder V2 model by downgrading from 0.2.27 to 0.2.25.

[barl0g]

Updating ollama is a solution for me:

curl -fsSL https://ollama.com/install.sh | sh

[jabeztadesse]

+1

[jabeztadesse]

I have the same error using, LM Studio 0.2.27, so it seems that this is an LMS problem.

Does it work in older versions? 🤔