-
Notifications
You must be signed in to change notification settings - Fork 0
/
Dockerfile
64 lines (50 loc) · 3.46 KB
/
Dockerfile
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
FROM nvcr.io/nvidia/pytorch:24.07-py3
RUN chown root:root /usr/lib
RUN apt update -y && apt install -y build-essential curl openssh-server openssh-client pdsh tmux
RUN pip install --upgrade pip wheel
RUN pip install \
accelerate \
deepspeed==0.15.1 \
openai \
msgspec \
peft \
pyarrow==14.0.2 \
sentencepiece \
tiktoken \
transformers \
trl
RUN pip install stanford-stk --no-deps
RUN pip uninstall -y pynvml
RUN pip install \
aioprometheus \
fastapi==0.111.0 \
fschat[model_worker,webui] \
gguf==0.9.1 \
lm-format-enforcer==0.10.6 \
outlines \
nvidia-ml-py \
prometheus-fastapi-instrumentator \
protobuf==3.20.3 \
ray==2.34.0 \
typer==0.12.3 \
uvicorn[standard]
RUN mkdir /packages/
ADD https://static.abacus.ai/pypi/abacusai/gh200-llm/pytorch-2407-cuda125/flash_attn-2.6.3-cp310-cp310-linux_aarch64.whl /packages/flash_attn-2.6.3-cp310-cp310-linux_aarch64.whl
ADD https://static.abacus.ai/pypi/abacusai/gh200-llm/pytorch-2407-cuda125/flash_attn-2.6.3-cp310-cp310-linux_x86_64.whl /packages/flash_attn-2.6.3-cp310-cp310-linux_x86_64.whl
RUN pip install --no-deps --find-links /packages flash-attn==2.6.3
ADD https://static.abacus.ai/pypi/abacusai/gh200-llm/pytorch-2407-cuda125/vllm_flash_attn-2.6.1%2Bcu125-cp310-cp310-linux_aarch64.whl /packages/vllm_flash_attn-2.6.1+cu125-cp310-cp310-linux_aarch64.whl
ADD https://static.abacus.ai/pypi/abacusai/gh200-llm/pytorch-2407-cuda125/vllm_flash_attn-2.6.1%2Bcu125-cp310-cp310-linux_x86_64.whl /packages/vllm_flash_attn-2.6.1+cu125-cp310-cp310-linux_x86_64.whl
RUN pip install --no-index --no-deps --find-links /packages vllm-flash-attn==2.6.1
ADD https://static.abacus.ai/pypi/abacusai/gh200-llm/pytorch-2407-cuda125/xformers-0.0.27.post2-cp310-cp310-linux_aarch64.whl /packages/xformers-0.0.27.post2-cp310-cp310-linux_aarch64.whl
ADD https://static.abacus.ai/pypi/abacusai/gh200-llm/pytorch-2407-cuda125/xformers-0.0.27.post2-cp310-cp310-linux_x86_64.whl /packages/xformers-0.0.27.post2-cp310-cp310-linux_x86_64.whl
RUN pip install --no-index --no-deps --find-links /packages xformers==0.0.27.post2
ADD https://static.abacus.ai/pypi/abacusai/gh200-llm/pytorch-2407-cuda125/megablocks-0.5.1-cp310-cp310-linux_aarch64.whl /packages/megablocks-0.5.1-cp310-cp310-linux_aarch64.whl
ADD https://static.abacus.ai/pypi/abacusai/gh200-llm/pytorch-2407-cuda125/megablocks-0.5.1-cp310-cp310-linux_x86_64.whl /packages/megablocks-0.5.1-cp310-cp310-linux_x86_64.whl
RUN pip install --no-deps --find-links /packages megablocks==0.5.1
ADD https://static.abacus.ai/pypi/abacusai/gh200-llm/pytorch-2407-cuda125/bitsandbytes-0.43.3-cp310-cp310-linux_aarch64.whl /packages/bitsandbytes-0.43.3-cp310-cp310-linux_aarch64.whl
ADD https://static.abacus.ai/pypi/abacusai/gh200-llm/pytorch-2407-cuda125/bitsandbytes-0.43.3-cp310-cp310-linux_x86_64.whl /packages/bitsandbytes-0.43.3-cp310-cp310-linux_x86_64.whl
RUN pip install --no-index --no-deps --find-links /packages bitsandbytes==0.43.3
ADD https://static.abacus.ai/pypi/abacusai/gh200-llm/pytorch-2407-cuda125/vllm-0.5.5%2Bcu125-cp310-cp310-linux_aarch64.whl /packages/vllm-0.5.5+cu125-cp310-cp310-linux_aarch64.whl
ADD https://static.abacus.ai/pypi/abacusai/gh200-llm/pytorch-2407-cuda125/vllm-0.5.5%2Bcu125-cp310-cp310-linux_x86_64.whl /packages/vllm-0.5.5+cu125-cp310-cp310-linux_x86_64.whl
RUN pip install --no-deps --find-links /packages vllm==0.5.5
RUN rm -r /packages