bitsandbytes==0.45.4
triton>=3.0.0
liger-kernel==0.5.5
packaging==23.2
peft==0.15.0
transformers==4.50.3
tokenizers>=0.21.1
accelerate==1.5.2
datasets==3.5.0
trl==0.16.0
optimum==1.16.2
hf_transfer
sentencepiece
gradio==3.50.2
modal==0.70.5
pydantic==2.10.6
addict
fire
PyYAML>=6.0
requests
wandb
einops
colorama
numba
numpy<=2.0.1,>=1.24.4
evaluate==0.4.1
scipy
scikit-learn==1.4.2
nvidia-ml-py==12.560.30
art
tensorboard
python-dotenv==1.0.1
s3fs>=2024.5.0
gcsfs>=2024.5.0
zstandard==0.22.0
fastcore
lm_eval==0.4.7
langdetect==1.0.9
immutabledict==4.2.0
antlr4-python3-runtime==4.13.2
torchao==0.7.0
schedulefree==1.3.0
axolotl-contribs-lgpl==0.0.6
axolotl-contribs-mit==0.0.3
torch==2.5.1
xformers>=0.0.28.post3

[apollo]
apollo-torch

[auto-gptq]
auto-gptq==0.5.1

[deepspeed]
deepspeed==0.15.4
deepspeed-kernels

[flash-attn]
flash-attn==2.7.4.post1

[galore]
galore_torch

[mamba-ssm]
mamba-ssm==1.2.0.post1
causal_conv1d

[mlflow]
mlflow

[optimizers]
galore_torch
apollo-torch
lomo-optim==0.1.1
torch-optimi==0.2.1

[ray]
ray[train]

[ring-flash-attn]
flash-attn==2.7.4.post1
ring-flash-attn>=0.1.4
yunchang==0.6.0

[vllm]
vllm==0.7.2
