[03/25, Paper] WaferLLM, the world fatest LLM inference system, has been accepted to OSDI 2025.
Emerging AI accelerators increasingly adopt wafer-scale manufacturing technologies, integrating hundreds of thousands of AI cores in a mesh-based architecture with large distributed on-chip memory (tens of GB in total) and ultra-high on-chip memory bandwidth (tens of PB/s). However, current LLM inference systems, optimized for shared memory architectures like GPUs, fail to fully exploit these accelerators.
We introduce WaferLLM, the first wafer-scale LLM inference system. WaferLLM is guided by a novel PLMR model (pronounced as “Plummer”) that captures the unique hardware characteristics of wafer-scale architectures. Leveraging this model, WaferLLM pioneers wafer-scale LLM parallelism, optimizing the utilization of hundreds of thousands of on-chip cores. It also introduces MeshGEMM and MeshGEMV, the first GEMM and GEMV implementations designed to scale effectively on wafer-scale accelerators.
Evaluations show that WaferLLM achieves 200× better wafer-scale accelerator utilization than state-of-the-art systems. On a commodity wafer-scale accelerator, WaferLLM delivers 606× faster and 22× more energy-efficient GEMV compared to an advanced GPU. For LLMs, based on 16-bit data type, WaferLLM achieves 2700 toks/sec/req decode speed on Llama3-8B model and 840 toks/sec/req decode speed on Qwen2-72B model, which enables 39× faster decoding with 1.7× better energy efficiency. We anticipate these numbers will grow significantly as wafer-scale AI models, software, and hardware continue to mature.