27 lines
635 B
Plaintext
27 lines
635 B
Plaintext
# ComfyUI-QwenVL requirements
|
|
# Note: Qwen3-VL models require transformers >= 4.57.0
|
|
|
|
transformers>=5.2.0
|
|
torch
|
|
huggingface-hub
|
|
hf_xet
|
|
psutil
|
|
numpy
|
|
Pillow
|
|
opencv-python
|
|
bitsandbytes
|
|
accelerate
|
|
sentencepiece
|
|
tiktoken
|
|
faster-whisper
|
|
|
|
# For running GGUF models with vision support
|
|
# The [server] extra installs necessary components for multimodal models
|
|
# llama-cpp-python[server]
|
|
|
|
# Optional: Flash Attention 2 for 2-3x performance boost (RTX 20+)
|
|
# flash-attn>=2.0.0 # Uncomment for Flash Attention 2 support
|
|
|
|
# Optional: SageAttention for 2-5x performance boost (RTX 30+)
|
|
# sageattention>=1.0.0 # Uncomment for SageAttention support
|