-Bench: Can Large Language Models Engineer the Infrastructure That Powers Them?

arXiv:2609.10226 · cs.CL · Submitted 2026-09-09 · Read on arXiv

cs.CL

Submitted: 2026-09-09

Updated: 2026-10-15

Code: https://github.com/NVIDIA/cutlass

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Benchmark — 2× in this paper · explained in 2 episodes
• Large Language Models (LLMs) — 1× in this paper · explained in 6 episodes

Sources

• Can Large Language Models Invent Algorithms to Improve Themselves?: Algorithm Discovery for Recursive Self-Improvement through Reinforcement Learning
• CUDAHercules: Benchmarking Hardware-Aware Expert-level CUDA Optimization for LLMs
• SOL-ExecBench: Speed-of-Light Benchmarking for Real-World GPU Kernels Against Hardware Limits
• TREX: Automating LLM Fine-tuning via Agent-Driven Tree-based Exploration
• ISO-Bench: Can Coding Agents Optimize Real-World Inference Workloads?
• KernelBench: Can LLMs Write Efficient GPU Kernels?
• Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism
• KernelBenchX: A Comprehensive Benchmark for Evaluating LLM-Generated GPU Kernels
• FlashInfer-Bench: Building the Virtuous Cycle for AI-driven LLM Systems
• GLM-5: from Vibe Coding to Agentic Engineering

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering