Huggingface_Hack / requirements.txt
minhahwang's picture
refactor: runtime config, Q&A flow module, storage paths, and cleanup
397b4ac
Raw
History Blame Contribute Delete
369 Bytes
gradio==5.35.0
transformers
torch
accelerate
bitsandbytes
soundfile
numpy
librosa
# LFM2.5-Audio end-to-end Q&A model
liquid-audio
# Qwen3-TTS (voice cloning + stock voice)
qwen-tts>=0.1.1
huggingface-hub>=0.23.0
# FlashAttention-2 is intentionally not installed at runtime on Spaces.
# The app falls back to SDPA unless a Docker/prebuilt image provides flash-attn.