Enable llama.cpp KV prompt cache (512 MB): ~21% off study-plan time

v2: Alpine base + runtime /admin/load (deepseek defaults + template kept)

Replace Gradio SmolVLM app with DeepSeek-7B-base GGUF OpenAI-compatible API (Docker)

Replace Gradio SmolVLM app with DeepSeek-7B-base GGUF OpenAI-compatible API (Docker)

Replace Gradio SmolVLM app with DeepSeek-7B-base GGUF OpenAI-compatible API (Docker)

Update app.py

Update app.py

Update requirements.txt

Update requirements.txt

Update app.py

Update requirements.txt

Create packages.txt

Update app.py

Update requirements.txt

Update app.py

Update requirements.txt

Update app.py

Update requirements.txt

Update app.py

Update requirements.txt

Update app.py

Update requirements.txt

Update app.py

Update requirements.txt

Update app.py

Update app.py

Update requirements.txt

Update app.py

Update requirements.txt

Update app.py

Create requirements.txt

Create app.py

initial commit