6 papers
SOLAR: AI-Powered Speed-of-Light Performance Analysis
Qijing Huang, Sana Damani, Zhifan Ye +9
How fast could a deep-learning model run on target hardware, and how far is today's implementation from that limit? These questions are central to software, hardware, and algorithm…
Regulating Branch Parallelism in LLM Serving
Swapnil Gandhi, Siva Hari, William J. Dally +1
Recent methods expose intra-request parallelism in LLM outputs, allowing independent branches to decode concurrently. Existing serving systems execute these branches eagerly or und…
Improving Efficiency of GPU Kernel Optimization Agents using a Domain-Specific Language and Speed-of-Light Guidance
Siva Kumar Sastry Hari, Vignesh Balaji, Sana Damani +2
Optimizing GPU kernels with LLM agents is an iterative process over a large design space. Every candidate must be generated, compiled, validated, and profiled, so fewer trials will…
SOL-ExecBench: Speed-of-Light Benchmarking for Real-World GPU Kernels Against Hardware Limits
Edward Lin, Sahil Modi, Siva Kumar Sastry Hari +30
As agentic AI systems become increasingly capable of generating and optimizing GPU kernels, progress is constrained by benchmarks that reward speedup over software baselines rather…
ProofWright: Towards Agentic Formal Verification of CUDA
Bodhisatwa Chatterjee, Drew Zagieboylo, Sana Damani +2
Large Language Models (LLMs) are increasingly used to automatically generate optimized CUDA kernels, substantially improving developer productivity. However, despite rapid generati…
KernelBlaster: Continual Cross-Task CUDA Optimization via Memory-Augmented In-Context Reinforcement Learning
Kris Shengjun Dong, Sahil Modi, Dima Nikiforov +4
Optimizing CUDA code across multiple generations of GPU architectures is challenging, as achieving peak performance requires an extensive exploration of an increasingly complex, ha…