Quantum adapters boost Llama 3.1 LLM performance on IBM's quantum hardware

Researchers have developed a method to enhance Large Language Models (LLMs) by integrating quantum circuit blocks, known as Cayley Unitary Adapters, into classical LLMs. Executed on an IBM Quantum System Two processor, this approach improved the perplexity of the Llama 3.1 8B model by 1.4%. A smaller model, SmolLM2, demonstrated that increasing the dimension of these quantum adapters leads to better performance, even enabling correct answers to questions that classical models failed. AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

IMPACT Demonstrates a potential pathway for quantum hardware to improve LLM performance beyond classical limitations.

RANK_REASON Academic paper detailing a novel method for integrating quantum computing with LLMs. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• Llama 3.1 8B
• IBM Quantum System Two
• SmolLM2
• Cayley Unitary Adapters

• paper
• infra