Describe the Issue
Llama.cpp already has support for the 3.0 moe, but kobold cant load with error 'llama_model_load: error loading model: unknown model architecture: 'bailingmoe3'. I'm simply launching with a ling tiny gguf.
Additional Information:
Linux, CPU inference, latest available build as of 24.08.26
Describe the Issue
Llama.cpp already has support for the 3.0 moe, but kobold cant load with error 'llama_model_load: error loading model: unknown model architecture: 'bailingmoe3'. I'm simply launching with a ling tiny gguf.
Additional Information:
Linux, CPU inference, latest available build as of 24.08.26