From cea31d16fb467d01e0c0c6202f1e8e86c33041c4 Mon Sep 17 00:00:00 2001 From: starkwj Date: Thu, 12 Feb 2026 11:13:26 +0800 Subject: [PATCH] add readme --- README.md | 229 ++++---------------------------------------------- README.md.bak | 223 ++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 240 insertions(+), 212 deletions(-) create mode 100644 README.md.bak diff --git a/README.md b/README.md index 778b707..af21dd4 100644 --- a/README.md +++ b/README.md @@ -1,223 +1,28 @@ -![vLLM Kunlun Logo](vllm_kunlun/patches/vLLM_Kunlun.jpg) +# XC-LLM: A Specially Optimized LLM Inference Engine for ModelHub XC -

- Documentation | - Quick Start | - Slack -

+## Overview ---- +The project is optimized based on the popular LLM inference project vLLM. The repository supports Kunlun3 P800. -## Latest News πŸ”₯ -- [2025/12] Initial release of vLLM Kunlun - ---- - -# Overview - -vLLM Kunlun (vllm-kunlun) is a community-maintained hardware plugin designed to seamlessly run vLLM on the Kunlun XPU. It is the recommended approach for integrating the Kunlun backend within the vLLM community, adhering to the principles outlined in the [RFC Hardware pluggable](https://github.com/vllm-project/vllm/issues/11162). This plugin provides a hardware-pluggable interface that decouples the integration of the Kunlun XPU with vLLM. - -By utilizing the vLLM Kunlun plugin, popular open-source models, including Transformer-like, Mixture-of-Expert, Embedding, and Multi-modal LLMs, can run effortlessly on the Kunlun XPU. - ---- -## Prerequisites - -- **Hardware**: Kunlun3 P800 -- **OS**: Ubuntu 22.04 -- **Software**: - - Python >=3.10 - - PyTorch β‰₯ 2.5.1 - - vLLM (same version as vllm-kunlun) - ---- -## Supported Models - -

Generaltive Models

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
ModelSupportQuantizationLoRAPiecewise Kunlun GraphNote
Qwen2βœ…βœ…βœ…
Qwen2.5βœ…βœ…βœ…
Qwen3βœ…βœ…βœ…
Qwen3-Moeβœ…βœ…βœ…βœ…
Qwen3-Nextβœ…βœ…βœ…βœ…
MiMo-V2-Flashβœ…βœ…
Llama2βœ…βœ…
Llama3βœ…βœ…
Llama3.1βœ…βœ…
gpt-ossβœ…
DeepSeek-R1βœ…βœ…βœ…
DeepSeek-V3βœ…βœ…βœ…
DeepSeek-V3.2βœ…βœ…βœ…
Kimi-K2βœ…βœ…βœ…
- -

Multimodal Language Models

- - - - - - - - - - - - - - - - - - - - - -
ModelSupportQuantizationLoRAPiecewise Kunlun GraphNote
Qwen3-VLβœ…βœ…
+One of the key features of this project is efficient memory coordination, enabling multiple vLLM instances share and dynamically hold GPU(XPU)'s physical memory. When an instance is idle, model parameters are offloaded to host memory. Upon a new inference request, the model parameters are quickly restored to the GPU’s memory (if not exist), without the need to initialize the engine and load the model from scratch. As a result, from the application’s perspective, multiple LLM inference engines can run on the GPU even when their total memory requirements exceed the physical memory limit. This technique is referred to as `InfiniVRAM`. +## Installation -## Performance Visualization πŸš€ -### High-performance computing at work: How different models perform on the Kunlun3 P800. +### Build from Dockerfile -Current environment: 16-way concurrency, input/output size 2048. +Clone this repository: +```bash +docker build -t $build_image -f ./Dockerfile . +``` -![Models and tgs](./vllm_kunlun/patches/performance.png) +## Usage -## Getting Started +0. To share GPU, processes coordinate via shm, so you need to set all containers with `ipc=host`. +1. Start a daemon process in a standalone container, by running `vllm_vxpu_daemon` installed inside the image. +2. Start LLM services with this image, following the official usage instructions. -Please use the following recommended versions to get started quickly: - -| Version | Release type | Doc | -|----------|---------------|-----| -| v0.11.0 | Latest stable version | [QuickStart](https://vllm-kunlun.readthedocs.io/en/latest/quick_start.html) and [Installation](https://vllm-kunlun.readthedocs.io/en/latest/installation.html) for more details | - ---- - -## Contribute to vLLM Kunlun - -If you're interested in contributing to this project, please read [Contributing](CONTRIBUTING.md) to vLLM Kunlun. - -## Star History πŸ”₯ - -We opened the project at Dec 8, 2025. We love open source and collaboration ❀️ - -[![Star History Chart](https://api.star-history.com/svg?repos=baidu/vLLM-Kunlun&type=date&legend=bottom-right)](https://www.star-history.com/#baidu/vLLM-Kunlun&type=date&legend=bottom-right) - -## Sponsors πŸ‘‹ - -We sincerely appreciate the [**KunLunXin**](https://www.kunlunxin.com/) team for their support in providing XPU resources, which enabled efficient model adaptation debugging, comprehensive end-to-end testing, and broader model compatibility. - -## License - -Apache License 2.0, as found in the [LICENSE](./LICENSE) file. +### Environment Variables +- `VXPU_RESERVED_VRAM_SIZE_GB`: The amonut of reserved GPU memory for other miscellaneous memory. Only needs to be set for `vllm_vxpu_daemon`. Try increasing the variable if you launch multiple LLM services and encounter OOM. Default: `8`. +- `VLLM_VXPU_SHM_NAME`: The name of the shm file. Needs to be set for all containers of the shared vxpu group. Default: `/vllm_kunlun_vxpu_offload_shm`. diff --git a/README.md.bak b/README.md.bak new file mode 100644 index 0000000..778b707 --- /dev/null +++ b/README.md.bak @@ -0,0 +1,223 @@ +![vLLM Kunlun Logo](vllm_kunlun/patches/vLLM_Kunlun.jpg) + +

+ Documentation | + Quick Start | + Slack +

+ +--- + +## Latest News πŸ”₯ +- [2025/12] Initial release of vLLM Kunlun + +--- + +# Overview + +vLLM Kunlun (vllm-kunlun) is a community-maintained hardware plugin designed to seamlessly run vLLM on the Kunlun XPU. It is the recommended approach for integrating the Kunlun backend within the vLLM community, adhering to the principles outlined in the [RFC Hardware pluggable](https://github.com/vllm-project/vllm/issues/11162). This plugin provides a hardware-pluggable interface that decouples the integration of the Kunlun XPU with vLLM. + +By utilizing the vLLM Kunlun plugin, popular open-source models, including Transformer-like, Mixture-of-Expert, Embedding, and Multi-modal LLMs, can run effortlessly on the Kunlun XPU. + +--- +## Prerequisites + +- **Hardware**: Kunlun3 P800 +- **OS**: Ubuntu 22.04 +- **Software**: + - Python >=3.10 + - PyTorch β‰₯ 2.5.1 + - vLLM (same version as vllm-kunlun) + +--- +## Supported Models + +

Generaltive Models

+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
ModelSupportQuantizationLoRAPiecewise Kunlun GraphNote
Qwen2βœ…βœ…βœ…
Qwen2.5βœ…βœ…βœ…
Qwen3βœ…βœ…βœ…
Qwen3-Moeβœ…βœ…βœ…βœ…
Qwen3-Nextβœ…βœ…βœ…βœ…
MiMo-V2-Flashβœ…βœ…
Llama2βœ…βœ…
Llama3βœ…βœ…
Llama3.1βœ…βœ…
gpt-ossβœ…
DeepSeek-R1βœ…βœ…βœ…
DeepSeek-V3βœ…βœ…βœ…
DeepSeek-V3.2βœ…βœ…βœ…
Kimi-K2βœ…βœ…βœ…
+ +

Multimodal Language Models

+ + + + + + + + + + + + + + + + + + + + + +
ModelSupportQuantizationLoRAPiecewise Kunlun GraphNote
Qwen3-VLβœ…βœ…
+ + + +## Performance Visualization πŸš€ +### High-performance computing at work: How different models perform on the Kunlun3 P800. + +Current environment: 16-way concurrency, input/output size 2048. + + +![Models and tgs](./vllm_kunlun/patches/performance.png) + +## Getting Started + +Please use the following recommended versions to get started quickly: + +| Version | Release type | Doc | +|----------|---------------|-----| +| v0.11.0 | Latest stable version | [QuickStart](https://vllm-kunlun.readthedocs.io/en/latest/quick_start.html) and [Installation](https://vllm-kunlun.readthedocs.io/en/latest/installation.html) for more details | + +--- + +## Contribute to vLLM Kunlun + +If you're interested in contributing to this project, please read [Contributing](CONTRIBUTING.md) to vLLM Kunlun. + +## Star History πŸ”₯ + +We opened the project at Dec 8, 2025. We love open source and collaboration ❀️ + +[![Star History Chart](https://api.star-history.com/svg?repos=baidu/vLLM-Kunlun&type=date&legend=bottom-right)](https://www.star-history.com/#baidu/vLLM-Kunlun&type=date&legend=bottom-right) + +## Sponsors πŸ‘‹ + +We sincerely appreciate the [**KunLunXin**](https://www.kunlunxin.com/) team for their support in providing XPU resources, which enabled efficient model adaptation debugging, comprehensive end-to-end testing, and broader model compatibility. + +## License + +Apache License 2.0, as found in the [LICENSE](./LICENSE) file.