LightCode: Compiling LLM Inference for Photonic-Electronic Systems
September 19, 2025 Β· Declared Dead Β· π arXiv.org
"No code URL or promise found in abstract"
Evidence collected by the PWNC Scanner
Authors
Ryan Tomich, Zhizhen Zhong, Dirk Englund
arXiv ID
2509.16443
Category
physics.app-ph
Cross-listed
cs.AI,
cs.PL
Citations
0
Venue
arXiv.org
Last Checked
3 months ago
Abstract
The growing demand for low-latency, energy-efficient inference in large language models (LLMs) has catalyzed interest in heterogeneous architectures. While GPUs remain dominant, they are poorly suited for integration with emerging domain-specific accelerators like the Photonic Tensor Units (PTUs), which offer low-power, high-throughput linear computation. This motivates hybrid compilation strategies that combine photonic and electronic resources. We present LightCode, a compiler framework and simulator for mapping LLM inference workloads across hybrid photonic-electronic systems. LightCode introduces the Stacked Graph, an intermediate representation that encodes multiple hardware-specific realizations of each tensor operation. Hardware assignment is formulated as a constrained subgraph selection problem optimized for latency or energy under parametric cost models. We evaluate LightCode on the prefill stage of GPT-2 and Llama-7B showing that under our workload and hardware assumptions, (i) Photonic hardware reduced energy by up to 50% in our simulated workloads at maximum sequence length; (ii) multiplexing and assignment strategy yielded latency improvements exceeding 10x; and (iii) Optimizing for latency or energy resulted in distinct hardware mappings in our simulations. LightCode offers a module, foundational framework and simulator for compiling LLMs to emerging photonic accelerators.
Community Contributions
Found the code? Know the venue? Think something is wrong? Let us know!
π Similar Papers
In the same crypt β physics.app-ph
R.I.P.
π»
Ghosted
R.I.P.
π»
Ghosted
Autonomous discovery of battery electrolytes with robotic experimentation and machine-learning
R.I.P.
π»
Ghosted
Harnessing The Multi-Stability Of Kresling Origami For Reconfigurable Articulation In Soft Robotic Arms
R.I.P.
π»
Ghosted
Deep learning for size-agnostic inverse design of random-network 3D printed mechanical metamaterials
R.I.P.
π»
Ghosted
Suction-based Soft Robotic Gripping of Rough and Irregular Parts
R.I.P.
π»
Ghosted
On-chip learning for domain wall synapse based Fully Connected Neural Network
Died the same way β π» Ghosted
R.I.P.
π»
Ghosted
Federated Learning: Strategies for Improving Communication Efficiency
R.I.P.
π»
Ghosted
In-Datacenter Performance Analysis of a Tensor Processing Unit
R.I.P.
π»
Ghosted
Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning
R.I.P.
π»
Ghosted