upd readme

README.md

@@ -13,3 +13,12 @@ pip install -r requirements.txt
 python -m spacy download ru_core_news_lg
 CMAKE_ARGS="-DGGML_CUDA=on -DLLAMA_OPENSSL=ON" pip install llama-cpp-python
 ```
+#### Сборка
+```
+cmake -B build -DGGML_CUDA=ON -DLLAMA_OPENSSL=ON
+cmake --build build --config Release -j$(nproc)
+```
+#### Запуск
+```
+llama-server -m models/vikhr-nemo-12b-instruct-r-21-09-24-q8_0.gguf --log-disable
+```