# Recurrent Drafter for Fast Speculative Decoding in Large Language Models

arXiv: https://arxiv.org/abs/2403.09919 · HTML: https://mlsanity.com/paper/2403.09919/

## Key facts

- Cited by 37 papers in the map (data through 14 Oct 2026).
- Placed in Speculative decoding, one of 12 directions in Efficient LLM inference and serving.
- Builds on 29 of its 42 references that are in the map.
- First posted on arXiv 14 Mar 2024.

## Placed in

- Efficient LLM inference and serving / [Speculative decoding](https://mlsanity.com/inference/speculative-decoding/)

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

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism](https://mlsanity.com/paper/1909.08053/) (2019)
- [Fast Inference from Transformers via Speculative Decoding](https://mlsanity.com/paper/2211.17192/) (2022)
- [PaLM 2 Technical Report](https://mlsanity.com/paper/2305.10403/) (2023)
- [Accelerating Large Language Model Decoding with Speculative Sampling](https://mlsanity.com/paper/2302.01318/) (2023)
- [Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads](https://mlsanity.com/paper/2401.10774/) (2024)
- [Sequence-Level Knowledge Distillation](https://mlsanity.com/paper/1606.07947/) (2016)
- [AlpacaFarm: A Simulation Framework for Methods that Learn from Human Feedback](https://mlsanity.com/paper/2305.14387/) (2023)
- [EAGLE: Speculative Sampling Requires Rethinking Feature Uncertainty](https://mlsanity.com/paper/2401.15077/) (2024)
- [Blockwise Parallel Decoding for Deep Autoregressive Models](https://mlsanity.com/paper/1811.03115/) (2018)
- [Break the Sequential Dependency of LLM Inference Using Lookahead Decoding](https://mlsanity.com/paper/2402.02057/) (2024)
- [Speculative Decoding: Exploiting Speculative Execution for Accelerating Seq2seq Generation](https://mlsanity.com/paper/2203.16487/) (2022)
- [DistillSpec: Improving Speculative Decoding via Knowledge Distillation](https://mlsanity.com/paper/2310.08461/) (2023)
- [Accelerating LLM Inference with Staged Speculative Decoding](https://mlsanity.com/paper/2308.04623/) (2023)
- [REST: Retrieval-Based Speculative Decoding](https://mlsanity.com/paper/2311.08252/) (2023)
- [Paraformer: Fast and Accurate Parallel Transformer for Non-autoregressive End-to-End Speech Recognition](https://mlsanity.com/paper/2206.08317/) (2022)
- [SpecTr: Fast Speculative Decoding via Optimal Transport](https://mlsanity.com/paper/2310.15141/) (2023)
- [Accelerating Transformer Inference for Translation via Parallel Decoding](https://mlsanity.com/paper/2305.10427/) (2023)
- [Truncation Sampling as Language Model Desmoothing](https://mlsanity.com/paper/2210.15191/) (2022)
- [Online Speculative Decoding](https://mlsanity.com/paper/2310.07177/) (2023)
- [Cascade Speculative Drafting for Even Faster LLM Inference](https://mlsanity.com/paper/2312.11462/) (2023)
- [Inference with Reference: Lossless Acceleration of Large Language Models](https://mlsanity.com/paper/2304.04487/) (2023)
- [Speculative Streaming: Fast LLM Inference without Auxiliary Models](https://mlsanity.com/paper/2402.11131/) (2024)
- [Amazon SageMaker Model Parallelism: A General and Flexible Framework for Large Model Training](https://mlsanity.com/paper/2111.05972/) (2021)

## Built on by (most cited first, 30 of 37)

- [SnapKV: LLM Knows What You are Looking for Before Generation](https://mlsanity.com/paper/2404.14469/) (2024)
- [EAGLE-2: Faster Inference of Language Models with Dynamic Draft Trees](https://mlsanity.com/paper/2406.16858/) (2024)
- [SpecDec++: Boosting Speculative Decoding via Adaptive Candidate Lengths](https://mlsanity.com/paper/2405.19715/) (2024)
- [Judge Decoding: Faster Speculative Sampling Requires Going Beyond Model Alignment](https://mlsanity.com/paper/2501.19309/) (2025)
- [Kangaroo: Lossless Self-Speculative Decoding via Double Early Exiting](https://mlsanity.com/paper/2404.18911/) (2024)
- [Speculative Diffusion Decoding: Accelerating Language Generation through Diffusion](https://mlsanity.com/paper/2408.05636/) (2024)
- [Your LLM Knows the Future: Uncovering Its Multi-Token Prediction Potential](https://mlsanity.com/paper/2507.11851/) (2025)
- [DySpec: Faster Speculative Decoding with Dynamic Token Tree Structure](https://mlsanity.com/paper/2410.11744/) (2024)
- [Speculative Decoding and Beyond: An In-Depth Survey of Techniques](https://mlsanity.com/paper/2502.19732/) (2025)
- [D-PACE: Dynamic Position-Aware Cross-Entropy for Parallel Speculative Drafting](https://mlsanity.com/paper/2605.18810/) (2026)
- [GRIFFIN: Effective Token Alignment for Faster Speculative Decoding](https://mlsanity.com/paper/2502.11018/) (2025)
- [Clover: Regressive Lightweight Speculative Decoding with Sequential Knowledge](https://mlsanity.com/paper/2405.00263/) (2024)
- [Bridging Draft Policy Misalignment: Group Tree Optimization for Speculative Decoding](https://mlsanity.com/paper/2509.22134/) (2025)
- [Adaptive Draft-Verification for Efficient Large Language Model Decoding](https://mlsanity.com/paper/2407.12021/) (2024)
- [Accelerating Production LLMs with Combined Token/Embedding Speculators](https://mlsanity.com/paper/2404.19124/) (2024)
- [FastDraft: How to Train Your Draft](https://mlsanity.com/paper/2411.11055/) (2024)
- [Clover-2: Accurate Inference for Regressive Lightweight Speculative Decoding](https://mlsanity.com/paper/2408.00264/) (2024)
- [A Survey on the Green Development of Large Models: From Resource-Efficient Architectures to Hardware-Software Co-Design](https://mlsanity.com/paper/2607.09084/) (2026)
- [TokenSwift: Lossless Acceleration of Ultra Long Sequence Generation](https://mlsanity.com/paper/2502.18890/) (2025)
- [WhisperKit: On-device Real-time ASR with Billion-Scale Transformers](https://mlsanity.com/paper/2507.10860/) (2025)
- [DeLS-Spec: Decoupled Long-Short Contexts for Parallel Speculative Drafting](https://mlsanity.com/paper/2607.07409/) (2026)
- [SpecBlock: Block-Iterative Speculative Decoding with Dynamic Tree Drafting](https://mlsanity.com/paper/2605.07243/) (2026)
- [SSSD: Simply-Scalable Speculative Decoding](https://mlsanity.com/paper/2411.05894/) (2024)
- [Hierarchical Verification of Speculative Beams for Accelerating LLM Inference](https://mlsanity.com/paper/2508.03726/) (2025)
- [TIDE: Temporal Incremental Draft Engine for Self-Improving LLM Inference](https://mlsanity.com/paper/2602.05145/) (2026)
- [MineDraft: A Framework for Batch Parallel Speculative Decoding](https://mlsanity.com/paper/2603.18016/) (2026)
- [H-Spec: Parallel Speculative Decoding Without a Drafter-Side KV Cache](https://mlsanity.com/paper/2609.24197/) (2026)
- [Cross-Attention Speculative Decoding](https://mlsanity.com/paper/2505.24544/) (2025)
- [Cross-Family Speculative Decoding for Polish Language Models on Apple~Silicon: An Empirical Evaluation of Bielik~11B with UAG-Extended MLX-LM](https://mlsanity.com/paper/2604.16368/) (2026)
- [EvoSpec: Evolving Speculative Decoding via Real-Time Vocabulary and Parameter Adaptation](https://mlsanity.com/paper/2605.27390/) (2026)

## Abstract (arXiv)

We present Recurrent Drafter (ReDrafter), an advanced speculative decoding approach that achieves state-of-the-art speedup for large language models (LLMs) inference. The performance gains are driven by three key aspects: (1) leveraging a recurrent neural network (RNN) as the draft model conditioning on LLM's hidden states, (2) applying a dynamic tree attention algorithm over beam search results to eliminate duplicated prefixes in candidate sequences, and (3) training through knowledge distillation from the LLM. ReDrafter accelerates Vicuna inference in MT-Bench by up to 2.8x with a PyTorch implementation on Nvidia H100 GPUs. To demonstrate its practicality in real environments, we also validated its effectiveness for on-device applications by implementing the approach in MLX and benchmarking performance on Metal GPUs in Apple Silicon chips, achieving up to 2.3x speedup.