testing llama-swap settings for performance

gistfile1.txt

#
# Run "write sname game in $lang" with different llama-swap profiles and llama.cpp
# greps logs for latest `eval time` 
#

for model in "qwen-coder-32b-q4-nodraft" "qwen-coder-32b-q4" "qwen-coder-32b-q4-w-ctk"; do 
    for lang in "python" "typescript" "swift"; do
        echo "Generating Snake Game in $lang using $model"
        curl -s --url http://localhost:8080/v1/chat/completions -d "{\"messages\": [{\"role\": \"system\", \"content\": \"you only write code.\"}, {\"role\": \"user\", \"content\": \"write snake game in $lang\"}], \"temperature\": 0.1, \"top_k\": 1, \"model\":\"$model\"}" > /dev/null
        curl -s --url http://localhost:8080/logs | grep 'eval time' | tail -n 2
        echo ""
    done
done
done