-
PH.D. student at Sun Yat-sen university
-
AI Infra, MLSys, Simulaters, GPU architecture
- [2025/06/27] [arXiv] [Code] gLLM is accepted by SC'25. Congratulations!
- [2025/05/28] [arXiv] [Code] EFIM is accepted by Euro-Par'25
- [2025/04/27] [arXiv] [Code] We have released gLLM, an efficient pipeline parallelism inference engine for LLM.
- SGLang: Support PP for zmq_to_scheduler link
- SGLang: Add EPD disaggregation doc link
- SGLang: feat: support EPD disaggregation link
- SGLang: [VLM] Support PP for Qwen2.5-VL link
- SGLang: Add PP support for dots_vlm link
- vLLM: [Bugfix] Fix benchmark_moe.py link
- SGLang: Fix port number overflow link
- xDiT: Enable warm up for VAE link
- xDiT: Fix parallel vae link
- DistVAE: Fix batch dimension link
- vLLM: [Benchmark] Refactor sample_requests in benchmark_throughput link
- vLLM: [Bugfix] fix automatic prefix args and add log info link
- vLLM: [Minor Fix] Fix comments in benchmark_serving link
- vLLM: [Minor Fix] Remove unused code in benchmark_prefix_caching.py link
- TVM: [Doc] Fix minor error in "Expressions in Relay" link
- TVM: [Doc] Fix minor error in doc (Add an operator to Relay) link