mirror of
https://github.com/huggingface/text-generation-inference.git
synced 2025-04-26 04:22:08 +00:00
a785000842
compressed-tensors is a safetensors extension for sparse, quantized tensors. The format is more powerful than earlier AWQ/GPTQ/FP8 quantization, because - Different quantizer configurations can be used for different targets. - The format can specify input/output quantizers in addition to weight quantizers. - Configurable exclusions for quantization. This change adds a dependency on the `compressed-tensors` package for its configuration parsing and layer matching functionality. The following types of quantization are supported in this PR: - W8A16 and W4A16 INT using GPTQ-Marlin kernels. - W8A8 and W8A16 FP using FP8-Marlin and cutlass kernels. Support for other quantization types will be added in subsequent PRs.
45 lines
1.6 KiB
Makefile
45 lines
1.6 KiB
Makefile
include Makefile-flash-att
|
|
include Makefile-flash-att-v2
|
|
include Makefile-vllm
|
|
include Makefile-awq
|
|
include Makefile-eetq
|
|
include Makefile-selective-scan
|
|
include Makefile-lorax-punica
|
|
include Makefile-exllamav2
|
|
include Makefile-flashinfer
|
|
|
|
unit-tests:
|
|
pytest -s -vv -m "not private" tests
|
|
|
|
gen-server:
|
|
# Compile protos
|
|
pip install grpcio-tools==1.62.2 mypy-protobuf==3.6.0 'types-protobuf' --no-cache-dir
|
|
mkdir text_generation_server/pb || true
|
|
python -m grpc_tools.protoc -I../proto/v3 --python_out=text_generation_server/pb \
|
|
--grpc_python_out=text_generation_server/pb --mypy_out=text_generation_server/pb ../proto/v3/generate.proto
|
|
find text_generation_server/pb/ -type f -name "*.py" -print0 -exec sed -i -e 's/^\(import.*pb2\)/from . \1/g' {} \;
|
|
touch text_generation_server/pb/__init__.py
|
|
|
|
install-server: gen-server
|
|
pip install pip --upgrade
|
|
pip install -r requirements_cuda.txt
|
|
pip install -e ".[accelerate, compressed-tensors, quantize, peft, outlines]"
|
|
|
|
|
|
install: install-cuda
|
|
echo "Installed server"
|
|
|
|
install-cuda: install-server install-flash-attention-v2-cuda install-vllm-cuda install-flash-attention
|
|
pip install -e ".[bnb,marlin,moe]"
|
|
pip install nvidia-nccl-cu12==2.22.3
|
|
|
|
install-rocm: install-server install-flash-attention-v2-rocm install-vllm-rocm
|
|
|
|
run-dev:
|
|
SAFETENSORS_FAST_GPU=1 python -m torch.distributed.run --nproc_per_node=2 text_generation_server/cli.py serve bigscience/bloom-560m --sharded
|
|
|
|
export-requirements:
|
|
poetry export -o requirements_cuda.txt --without-hashes
|
|
poetry export -o requirements_rocm.txt --without-hashes
|
|
poetry export -o requirements_intel.txt --without-hashes
|