fix: use python3.10 instead of python3.9 for ubuntu 22.04

fix: defer model downloads to first request

refactor: defer llama-cpp-python install to runtime

fix: use symlinks instead of update-alternatives for python

fix: correct Dockerfile syntax for llama-cpp-python fallback

refactor: consolidate to single unified Dockerfile with GPU support

feat: add GPU-enabled Dockerfile.gpu for HF Spaces CUDA support

fix: graceful fallback for llama-cpp-python installation on HF Spaces

fix: enable CUDA compilation for llama-cpp-python

perf: defer llama-cpp-python build to runtime startup

fix: remove invalid chown command from Dockerfile

update Dockerfile and README.md to replace Qwen2.5-3B and Gemma-2-2B with Bielik-1.5B-GGUF; adjust model loading instructions in the API documentation

dockerfile fix

fixing naming for bielik gguf

fix-docker-error-for-gguf

adding-bielik-gguf

pre-downloading-all-models-at-startup

Feat: Simplify model download using huggingface-cli

update dockerfile with appropriate token fetch

update_dockerfile

adding-github-files-to-spaces