# Diffusion LLM and MoE inference

Area: [Efficient LLM inference and serving](https://mlsanity.com/inference/index.md). HTML: https://mlsanity.com/inference/diffusion-llm-and-moe-inference/

## Key facts

- Diffusion LLM and MoE inference: 156 papers on arXiv since 2024 (data through 14 Oct 2026).
- 76 of its 156 mapped papers (49%) cite “Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding” (2025, arXiv 2505.22618).
- 39 of its 156 mapped papers (25%) cite “dKV-Cache: The Cache for Diffusion Language Models” (2025, arXiv 2505.15781).
- 14 new papers 24 Sep–14 Oct.

## Most cited here

- [Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding](https://mlsanity.com/paper/2505.22618/) (2025), cited by 76 papers here
- [dKV-Cache: The Cache for Diffusion Language Models](https://mlsanity.com/paper/2505.15781/) (2025), cited by 39 papers here
- [Accelerating Diffusion LLMs via Adaptive Parallel Decoding](https://mlsanity.com/paper/2506.00413/) (2025), cited by 27 papers here
- [Diffusion LLMs Can Do Faster-Than-AR Inference via Discrete Diffusion Forcing](https://mlsanity.com/paper/2508.09192/) (2025), cited by 27 papers here
- [dParallel: Learnable Parallel Decoding for dLLMs](https://mlsanity.com/paper/2509.26488/) (2025), cited by 23 papers here
- [Sparse-dLLM: Accelerating Diffusion LLMs with Dynamic Cache Eviction](https://mlsanity.com/paper/2508.02558/) (2025), cited by 20 papers here
- [Dimple: Discrete Diffusion Multimodal Large Language Model with Parallel Decoding](https://mlsanity.com/paper/2505.16990/) (2025), cited by 17 papers here
- [ParallelBench: Understanding the Trade-offs of Parallel Decoding in Diffusion LLMs](https://mlsanity.com/paper/2510.04767/) (2025), cited by 13 papers here
- [Learning to Parallel: Accelerating Diffusion Large Language Models via Learnable Parallel Decoding](https://mlsanity.com/paper/2509.25188/) (2025), cited by 12 papers here
- [Free Draft-and-Verification: Toward Lossless Parallel Decoding for Diffusion Large Language Models](https://mlsanity.com/paper/2510.00294/) (2025), cited by 11 papers here

## New (24 Sep–14 Oct)

- [Energy Consumption Analysis of Discrete Diffusion and Autoregressive Language Models](https://mlsanity.com/paper/2610.19855/) (2026-10-14)
- [Revisiting the 16x Pretraining Compute Gap between Diffusion and Autoregressive Language Models](https://mlsanity.com/paper/2610.17590/) (2026-10-13)
- [Intent-Triggered Anchoring for Multi-Token Format Units in Parallel Decoding of Diffusion Language Models](https://mlsanity.com/paper/2610.16516/) (2026-10-12)
- [Enhancing Diffusion Language Models with Autoregressive Post-Training Weights](https://mlsanity.com/paper/2610.08108/) (2026-10-06)
- [SoloQ: Calibration-Free Quantization for Diffusion Language Models](https://mlsanity.com/paper/2610.07121/) (2026-10-05)
- [SpecFold: Folding Multi-Branch Redundancy for Faster Speculative Decoding in Diffusion Language Models](https://mlsanity.com/paper/2610.04875/) (2026-10-04)
- [Large Language Continuous Diffusion Models](https://mlsanity.com/paper/2610.02665/) (2026-10-02)
- [ITC-MoE: Importance-guided Token-aware Compression for MoE Diffusion Language Models](https://mlsanity.com/paper/2610.01296/) (2026-10-01)
- [Clock Diffusion: Efficient Semi-Autoregressive Continuous Diffusion Language Models](https://mlsanity.com/paper/2610.00894/) (2026-10-01)
- [Rank-Aware Speculative Sampling for Diffusion Draft Trees](https://mlsanity.com/paper/2610.02251/) (2026-09-30)
- [Denoising Surface: Modeling and Predicting Inference Cost for Diffusion LLM Serving](https://mlsanity.com/paper/2610.00499/) (2026-09-30)
- [Does This Action Still Explain the Task? Reverse Scoring for Diffusion Language Model Agents](https://mlsanity.com/paper/2609.38536/) (2026-09-29)
- [Rethinking Soft Tokens for Parallel Decoding in Diffusion Language Models](https://mlsanity.com/paper/2609.37391/) (2026-09-29)
- [Reliable Parallel Decoding in Masked Diffusion Language Models](https://mlsanity.com/paper/2609.36452/) (2026-09-29)