SOLAR: AI-Powered Speed-of-Light Performance Analysis
Not on the map. Why?
Abstract (arXiv)
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 optimizations. Speed-of-Light (SOL) analysis answers them by computing a workload's theoretical minimum execution time on a given architecture. Yet deriving SOL bounds remains manual, error-prone, and disconnected from rapid model development. To close this gap, we introduce SOLAR, a framework that automatically derives validated SOL bounds from PyTorch and JAX source code. SOLAR leverages both generative and deterministic components in its flow: an LLM frontend translates any source programs into an executable Affine Loop IR, validated by output comparison; a deterministic flow lifts the IR into an einsum graph; and an analytical backend computes unfused, fused, and cache-aware SOL bounds. SOLAR provides comprehensive operator and language coverage, produces validated bounds with zero observed SOL violations, and offers multi-fidelity analysis that tightens bounds and surfaces optimization insights. We evaluate SOLAR across KernelBench, JAX/Flax models, and robotics workloads. These experiments demonstrate four use cases: headroom analysis at multiple fidelity levels, identifying optimization opportunities, cross-platform exploration, and inverse-roofline hardware provisioning.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 12 of its 23 references that are in the map. First posted on arXiv 24 Jun 2026.
Builds on
12 of 23 references.
References · 12
- 2017Attention Is All You Needcited by 60,361
- 2021Evaluating Large Language Models Trained on Codecited by 8,320
- 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,276
- 2023SWE-bench: Can Language Models Resolve Real-World GitHub Issues?cited by 3,373
- 2022Competition-Level Code Generation with AlphaCodecited by 1,462
- 2024Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Dualitycited by 1,336
- 2025KernelBench: Can LLMs Write Efficient GPU Kernels?cited by 225
- 2021A Runtime-Based Computational Performance Predictor for Deep Neural Network Trainingcited by 29
- 2026SOL-ExecBench: Speed-of-Light Benchmarking for Real-World GPU Kernels Against Hardware Limitscited by 18
- 2026Improving Efficiency of GPU Kernel Optimization Agents using a Domain-Specific Language and Speed-of-Light Guidancecited by 5
- 2025ASAP: an Agentic Solution to Auto-optimize Performance of Large-Scale LLM Trainingcited by 3
- 2024The EDGE Language: Extended General Einsums for Graph Algorithmscited by 2
Built on by
Not cited yet.