Compile bug: ./llama-server: symbol lookup error: ./llama-server: undefined symbol: llama_vocab_eos

[sakshi-joshi-handle]

Git commit

73e2ed3ce3492d3ed70193dd09ae8aa44779651d

Operating systems

Linux

GGML backends

CUDA

Problem description & steps to reproduce

I am trying to host model using llama-server. Successfully built llama.cpp package but getting error when running llama-server.

./build/bin/llama-server: symbol lookup error: ./build/bin/llama-server: undefined symbol: llama_vocab_eos

Exact same error while running model via llama-cli - ./build/bin/llama-cli: symbol lookup error: ./build/bin/llama-cli: undefined symbol: llama_vocab_eos

First Bad Commit

No response

Compile command

./build/bin/llama-server \
    -m /path/to/Meta-Llama-3.1-8B-Instruct-GGUF/Meta-Llama-3.1-8B-Instruct-Q4_K_M.gguf \
    --threads 6 \
    --n-gpu-layers 1000 \
    --ctx-size 8192 \
    --host 0.0.0.0 \
    --port 8080 \
    -fa

Relevant log output

[ 79%] Building CXX object examples/main/CMakeFiles/llama-cli.dir/main.cpp.o
[ 80%] Linking CXX executable ../../bin/llama-cli
[ 80%] Built target llama-cli

[ 84%] Building CXX object examples/server/CMakeFiles/llama-server.dir/server.cpp.o
[ 85%] Linking CXX executable ../../bin/llama-server
[ 85%] Built target llama-server