accelerate
codetiming
datasets
# flash-attn>=2.4.3
liger-kernel
mathruler
numpy
omegaconf
pandas
peft
pillow
pyarrow>=15.0.0
pylatexenc
qwen-vl-utils
ray[default]
tensordict
torchdata
# transformers>=4.54.0,<=4.57.0
# vllm>=0.8.0 
wandb