masx-openchat-llm / requirements.txt
ateetvatan
swiched to 4-bit AWQ‑quantized version of OpenChat 3.5 7B as openchat 3.5 requires more then 16gn RAM
62d49a1
raw
history blame contribute delete
172 Bytes
# Core API
fastapi>=0.104.0
uvicorn[standard]>=0.24.0
pydantic>=2.5.0
# Quantized LLM support
ctransformers>=0.2.27
# Optional: For local .env files
python-dotenv>=1.0.0