Skip to content
View david6666666's full-sized avatar
💭
holiady
💭
holiady

Block or report david6666666

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
david6666666/README.md
Building efficient multimodal inference — text, image, audio and video

WeiQing Chen · vLLM-Omni Committer · 专注多模态与大模型推理:模型接入、分布式服务、GPU/NPU 性能优化,以及可复现的基准与质量验证。

🔭 Focus

  • 多模态推理:在 vLLM-Omni 中接入与优化图像、音频、视频等模型,打通模型加载与推理服务
  • 性能优化:从 profiling 定位瓶颈,优化 GPU kernel、通信与执行流程,同时验证输出质量
  • 分布式服务:关注专家并行负载均衡、异步专家重排,以及多模态投机解码
  • Agent 辅助优化:通过 InferRSI 组织基准对比、质量检查与实验记录,让优化过程可追溯、经验可复用

🧩 Featured Projects

项目 说明 技术
vLLM-Omni Committer;面向文本、图像、音频与视频模型的高效推理框架,参与模型接入、性能优化与质量修复 PyTorch · Diffusion · GPU/NPU
InferRSI 参与构建面向 coding agent 的推理优化工作台:组织受控基准对比、质量检查、实验报告与可复用调优知识 Python · Benchmarking · Agent Skills
AgentInfer 面向 Agent 工作负载的路由、调度与缓存框架;参与 InferRSI 源码集成 vLLM · KV Cache · Agent Workloads

🌱 Open Source

vLLM-Omni Committer,参与 vLLM 与 AgentInfer 开源协作。部分已合并贡献:

🛠 Tech Stack

Python, PyTorch, Kubernetes, Git, Linux

Inference:vLLM · vLLM-Omni · PyTorch Infrastructure:CUDA · Ascend NPU · Ray · Kubernetes

🐍 Contributions

Animated snake tracing david6666666's GitHub contributions

Pinned Loading

  1. vllm-project/vllm vllm-project/vllm Public

    A high-throughput and memory-efficient inference and serving engine for LLMs

    Python 96.2k 24.3k

  2. vllm-project/vllm-omni vllm-project/vllm-omni Public

    A framework for efficient model inference with omni-modality models

    Python 7.5k 2k

  3. InferMatrixLab/InferRSI InferMatrixLab/InferRSI Public

    Python 56 12