Failed to load model
🥲 加载模型失败
Failed to load model
error loading model: error loading model architecture: unknown model architecture: 'ernie4_5'
You'll need to update llama.cpp from source! For example in our Gemma 3N blog post https://docs.unsloth.ai/basics/gemma-3n-how-to-run-and-fine-tune:
apt-get update
apt-get install pciutils build-essential cmake curl libcurl4-openssl-dev -y
git clone https://github.com/ggerganov/llama.cpp
cmake llama.cpp -B llama.cpp/build \
-DBUILD_SHARED_LIBS=ON -DGGML_CUDA=ON -DLLAMA_CURL=ON
cmake --build llama.cpp/build --config Release -j --clean-first --target llama-quantize llama-cli llama-gguf-split llama-mtmd-cli
cp llama.cpp/build/bin/llama-* llama.cpp
Hi, am using llama.cpp 5835 and got this error with llama-server: nvm, I am stupid (adding --jinja fixes it)
srv load_model: load_model: Chat template parsing error: this custom template is not supported, try using --jinja srv load_model: load_model: The chat template that comes with this model is not yet supported, falling back to chatml. This may cause the model to output suboptimal responses
and this with llama-cli:main: chat template is available, enabling conversation mode (disable it with -no-cnv) /bin/libggml-base.so(+0x15bcc) [0x7f974914dbcc] /bin/libggml-base.so(ggml_print_backtrace+0x250) [0x7f974914e070] /bin/libggml-base.so(+0x26d40) [0x7f974915ed40] /usr/lib/libstdc++.so.6(+0xb1c1a) [0x7f9748cb1c1a] /usr/lib/libstdc++.so.6(_ZSt10unexpectedv+0x0) [0x7f9748c975db] /usr/lib/libstdc++.so.6(+0xb1ed6) [0x7f9748cb1ed6] ./llama-cli(+0x34780) [0x55cc1ff2b780] ./llama-cli(+0xd5ce7) [0x55cc1ffccce7] ./llama-cli(+0x3ebfb) [0x55cc1ff35bfb] /usr/lib/libc.so.6(+0x276b5) [0x7f9748a376b5] /usr/lib/libc.so.6(__libc_start_main+0x89) [0x7f9748a37769] ./llama-cli(+0x425b5) [0x55cc1ff395b5] terminate called after throwing an instance of 'std::runtime_error' what(): this custom template is not supported, try using --jinja Aborted (core dumped)