Add markdown rendering (marked.js + highlight.js)

Add chat UI (mirror 27B space)

Switch to Q8_K_P, add chat.html

Update MODEL_PATH to Q8_K_P, add threading + reasoning flags

Update proxy to match 27B space (chat UI, hop-by-hop headers, api-info)

Update title to Q8, update README

Use wget instead of deprecated huggingface-cli

Fix pip install with --break-system-packages for Ubuntu 24.04

Set app_port to 8000

Use local pre-downloaded model and start uvicorn proxy

Robust proxy with lifespan-based llama-server readiness check

Pre-download model in Dockerfile and fix proxy startup

Upload Dockerfile

Upload Dockerfile

Upload Dockerfile

Upload entrypoint.sh

Upload README.md

Upload proxy.py

Upload Dockerfile

Update to 131K context with parallel 1

Upload README.md

Fix Dockerfile - use ENTRYPOINT with correct binary path

Add README with API docs

Add Dockerfile for llama.cpp server

initial commit