# SlimWise: Decoupling Expert Pruning Across Prefill and Decode for Efficient MoE Serving

arXiv: https://arxiv.org/abs/2609.34117 · HTML: https://mlsanity.com/paper/2609.34117/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Efficient MoE training and inference systems, one of 12 directions in Mixture-of-experts.
- Builds on 29 of its 36 references that are in the map.
- First posted on arXiv 28 Sep 2026.

## Placed in

- Mixture-of-experts / [Efficient MoE training and inference systems](https://mlsanity.com/moe/efficient-moe-training-and-inference-systems/)
- Efficient LLM inference and serving / [Early exit and edge inference](https://mlsanity.com/inference/early-exit-and-edge-inference/)

## Builds on (most cited first, 29 of 29)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [Let's Verify Step by Step](https://mlsanity.com/paper/2305.20050/) (2023)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [Mixtral of Experts](https://mlsanity.com/paper/2401.04088/) (2024)
- [Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generation](https://mlsanity.com/paper/2305.01210/) (2023)
- [Splitwise: Efficient generative LLM inference using phase splitting](https://mlsanity.com/paper/2311.18677/) (2023)
- [Taming Throughput-Latency Tradeoff in LLM Inference with Sarathi-Serve](https://mlsanity.com/paper/2403.02310/) (2024)
- [Gemma 4 Technical Report](https://mlsanity.com/paper/2607.02770/) (2026)
- [Mooncake: A KVCache-centric Disaggregated Architecture for LLM Serving](https://mlsanity.com/paper/2407.00079/) (2024)
- [APIGen-MT: Agentic Pipeline for Multi-Turn Data Generation via Simulated Agent-Human Interplay](https://mlsanity.com/paper/2504.03601/) (2025)
- [Merge, Then Compress: Demystify Efficient SMoE with Hints from Its Routing Policy](https://mlsanity.com/paper/2310.01334/) (2023)
- [NVIDIA Nemotron Nano 2: An Accurate and Efficient Hybrid Mamba-Transformer Reasoning Model](https://mlsanity.com/paper/2508.14444/) (2025)
- [MoE-Pruner: Pruning Mixture-of-Experts Large Language Model using the Hints from Its Router](https://mlsanity.com/paper/2410.12013/) (2024)
- [REAP the Experts: Why Pruning Prevails for One-Shot MoE compression](https://mlsanity.com/paper/2510.13999/) (2025)
- [Retraining-Free Merging of Sparse MoE via Hierarchical Clustering](https://mlsanity.com/paper/2410.08589/) (2024)
- [Finding Fantastic Experts in MoEs: A Unified Study for Expert Dropping Strategies and Observations](https://mlsanity.com/paper/2504.05586/) (2025)
- [Lynx: Enabling Efficient MoE Inference through Dynamic Batch-Aware Expert Selection](https://mlsanity.com/paper/2411.08982/) (2024)
- [Quantized Reasoning Models Think They Need to Think Longer, but They Do Not](https://mlsanity.com/paper/2606.00206/) (2026)
- [Janus: Disaggregating Attention and Experts for Scalable MoE Inference](https://mlsanity.com/paper/2512.13525/) (2025)
- [Opportunistic Expert Activation: Batch-Aware Expert Routing for Faster Decode Without Retraining](https://mlsanity.com/paper/2511.02237/) (2025)
- [XShare: Collaborative in-Batch Expert Sharing for Faster MoE Inference](https://mlsanity.com/paper/2602.07265/) (2026)
- [SlimQwen: Exploring the Pruning and Distillation in Large MoE Model Pre-training](https://mlsanity.com/paper/2605.08738/) (2026)
- [ELDR: Expert-Locality-Aware Decode Routing for PD-Disaggregated MoE Serving](https://mlsanity.com/paper/2607.00466/) (2026)
- [ExFold: Unified Expert Folding for Training-Free MoE Prefill-Decode Acceleration](https://mlsanity.com/paper/2608.24938/) (2026)
- [Pruning and Distilling Mixture-of-Experts into Dense Language Models](https://mlsanity.com/paper/2605.28207/) (2026)

## Abstract (arXiv)

Mixture-of-experts (MoE) models activate few experts per token, yet batched decoding can access nearly the entire expert pool, making expert-weight traffic a major bottleneck. Expert pruning reduces this traffic, but conventional approaches also prune compute-bound prefill, sacrificing model quality for little throughput benefit. We present SlimWise, a serving framework that tailors the expert pool to each inference phase. SlimWise performs prefill with the full model and decode with a pruned model that directly reuses the prefill-generated KV cache without conversion. Across two MoE backbones and three pruning criteria, this training-free KV cache handoff substantially narrows accuracy gaps relative to the full model in many settings. We also show that benchmark accuracy can conceal substantial pruning-induced changes in generation length. To address these distortions and residual accuracy loss, SlimWise introduces a low-cost distillation stage that trains the decoder to continue from full-model KV caches while updating only a small subset of parameters. Implemented in vLLM, SlimWise supports both prefill-decode (PD) disaggregation and PD-colocated serving. On Qwen3.6-35B-A3B, SlimWise improves decode throughput by up to 1.81x at 50% expert pruning with minimal accuracy loss.