mirror of
https://github.com/kvcache-ai/ktransformers.git
synced 2026-10-17 05:11:19 +08:00
Jiaqi Liao
57d14d22bc
* refactor: move legacy code to archive/ directory - Moved ktransformers, csrc, third_party, merge_tensors to archive/ - Moved build scripts and configurations to archive/ - Kept kt-kernel, KT-SFT, doc, and README files in root - Preserved complete git history for all moved files * refactor: restructure repository to focus on kt-kernel and KT-SFT modules * fix README * fix README * fix README * fix README * docs: add performance benchmarks to kt-kernel section Add comprehensive performance data for kt-kernel to match KT-SFT's presentation: - AMX kernel optimization: 21.3 TFLOPS (3.9× faster than PyTorch) - Prefill phase: up to 20× speedup vs baseline - Decode phase: up to 4× speedup - NUMA optimization: up to 63% throughput improvement - Multi-GPU (8×L20): 227.85 tokens/s total throughput with DeepSeek-R1 FP8 Source: https://lmsys.org/blog/2025-10-22-KTransformers/ This provides users with concrete performance metrics for both core modules, making it easier to understand the capabilities of each component. * refactor: improve kt-kernel performance data with specific hardware and models Replace generic performance descriptions with concrete benchmarks: - Specify exact hardware: 8×L20 GPU + Xeon Gold 6454S, Single/Dual-socket Xeon + AMX - Include specific models: DeepSeek-R1-0528 (FP8), DeepSeek-V3 (671B) - Show detailed metrics: total throughput, output throughput, concurrency details - Match KT-SFT presentation style for consistency This provides users with actionable performance data they can use to evaluate hardware requirements and expected performance for their use cases. * fix README * docs: clean up performance table and improve formatting * add pic for README * refactor: simplify .gitmodules and backup legacy submodules - Remove 7 legacy submodules from root .gitmodules (archive/third_party/*) - Keep only 2 active submodules for kt-kernel (llama.cpp, pybind11) - Backup complete .gitmodules to archive/.gitmodules - Add documentation in archive/README.md for researchers who need legacy submodules This reduces initial clone size by ~500MB and avoids downloading unused dependencies. * refactor: move doc/ back to root directory Keep documentation in root for easier access and maintenance. * refactor: consolidate all images to doc/assets/ - Move kt-kernel/assets/heterogeneous_computing.png to doc/assets/ - Remove KT-SFT/assets/ (images already in doc/assets/) - Update KT-SFT/README.md image references to ../doc/assets/ - Eliminates ~7.9MB image duplication - Centralizes all documentation assets in one location * fix pic path for README
64 lines
1.4 KiB
Docker
64 lines
1.4 KiB
Docker
FROM pytorch/pytorch:2.5.1-cuda12.1-cudnn9-devel as compile_server
|
|
|
|
|
|
ARG CPU_INSTRUCT=NATIVE
|
|
|
|
# 设置工作目录和 CUDA 路径
|
|
WORKDIR /workspace
|
|
ENV CUDA_HOME=/usr/local/cuda
|
|
|
|
|
|
|
|
# 安装依赖
|
|
RUN apt update -y
|
|
RUN apt install -y --no-install-recommends \
|
|
libtbb-dev \
|
|
libssl-dev \
|
|
libcurl4-openssl-dev \
|
|
libaio1 \
|
|
libaio-dev \
|
|
libfmt-dev \
|
|
libgflags-dev \
|
|
zlib1g-dev \
|
|
patchelf \
|
|
git \
|
|
wget \
|
|
vim \
|
|
gcc \
|
|
g++ \
|
|
cmake
|
|
# 拷贝代码
|
|
RUN git clone https://github.com/kvcache-ai/ktransformers.git
|
|
# 清理 apt 缓存
|
|
RUN rm -rf /var/lib/apt/lists/*
|
|
|
|
# 进入项目目录
|
|
WORKDIR /workspace/ktransformers
|
|
# 初始化子模块
|
|
RUN git submodule update --init --recursive
|
|
|
|
# 升级 pip
|
|
RUN pip install --upgrade pip
|
|
|
|
# 安装构建依赖
|
|
RUN pip install ninja pyproject numpy cpufeature aiohttp zmq openai
|
|
|
|
# 安装 flash-attn（提前装可以避免后续某些编译依赖出错）
|
|
RUN pip install flash-attn
|
|
|
|
# 安装 ktransformers 本体（含编译）
|
|
RUN CPU_INSTRUCT=${CPU_INSTRUCT} \
|
|
USE_BALANCE_SERVE=1 \
|
|
KTRANSFORMERS_FORCE_BUILD=TRUE \
|
|
TORCH_CUDA_ARCH_LIST="8.0;8.6;8.7;8.9;9.0+PTX" \
|
|
pip install . --no-build-isolation --verbose
|
|
|
|
RUN pip install third_party/custom_flashinfer/
|
|
# 清理 pip 缓存
|
|
RUN pip cache purge
|
|
|
|
# 拷贝 C++ 运行时库
|
|
RUN cp /usr/lib/x86_64-linux-gnu/libstdc++.so.6 /opt/conda/lib/
|
|
|
|
# 保持容器运行（调试用）
|
|
ENTRYPOINT ["tail", "-f", "/dev/null"] |