Popular repositories Loading

1. nanoinfer-lab nanoinfer-lab Public

   A learning-oriented single-GPU LLM inference runtime with reproducible RTX 5060 experiments.

   Python 1

2. cpp-live-commerce-flash-sale-backend cpp-live-commerce-flash-sale-backend Public

   C++ 直播电商秒杀消费后台：基于 C++17、Reactor、Redis、RabbitMQ 与 MySQL 的高并发秒杀和可靠异步订单系统

   C++ 1

3. Mymoduo Mymoduo Public

   C++11 Reactor-style TCP networking library inspired by muduo, with epoll, eventfd, thread pool, CI, and docs.

   C++

4. ConcurrentCacheSystem ConcurrentCacheSystem Public

   C++ concurrent cache system with LRU/LFU/ARC policies, tests, benchmarks, and CI.

   C++

5. nano-vllm-sm120-optimized nano-vllm-sm120-optimized Public

   面向 RTX 5060 Blackwell SM120 的离线 LLM 推理引擎：Paged Attention、KV Cache、连续批处理、CUDA Graph 与算子融合

   Python

6. sm120-norm-fusion-kernels sm120-norm-fusion-kernels Public

   面向 Blackwell SM120 的 LLM Norm 融合算子：RMSNorm、RoPE、Paged KV Store 与 NCU/SASS 性能验证

   Python