# An Efficient vLLM-Based Inference Pipeline for Unified Audio Understanding and Generation

arXiv: https://arxiv.org/abs/2607.02119 · HTML: https://mlsanity.com/paper/2607.02119/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Video understanding with MLLMs, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 30 of its 32 references that are in the map.
- First posted on arXiv 2 Jul 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Video understanding with MLLMs](https://mlsanity.com/mllm/video-understanding-with-mllms/)

## Builds on (most cited first, 30 of 30)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [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)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [SGLang: Efficient Execution of Structured Language Model Programs](https://mlsanity.com/paper/2312.07104/) (2023)
- [Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions](https://mlsanity.com/paper/1712.05884/) (2017)
- [AudioLDM: Text-to-Audio Generation with Latent Diffusion Models](https://mlsanity.com/paper/2301.12503/) (2023)
- [Moshi: a speech-text foundation model for real-time dialogue](https://mlsanity.com/paper/2410.00037/) (2024)
- [Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speech](https://mlsanity.com/paper/2106.06103/) (2021)
- [FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision](https://mlsanity.com/paper/2407.08608/) (2024)
- [Qwen3-Omni Technical Report](https://mlsanity.com/paper/2509.17765/) (2025)
- [Simple and Controllable Music Generation](https://mlsanity.com/paper/2306.05284/) (2023)
- [High-Fidelity Audio Compression with Improved RVQGAN](https://mlsanity.com/paper/2306.06546/) (2023)
- [2 OLMo 2 Furious](https://mlsanity.com/paper/2501.00656/) (2024)
- [SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model](https://mlsanity.com/paper/2502.02737/) (2025)
- [MMAU: A Massive Multi-Task Audio Understanding and Reasoning Benchmark](https://mlsanity.com/paper/2410.19168/) (2024)
- [Kimi-Audio Technical Report](https://mlsanity.com/paper/2504.18425/) (2025)
- [Codec Does Matter: Exploring the Semantic Shortcoming of Codec for Audio Language Model](https://mlsanity.com/paper/2408.17175/) (2024)
- [Understanding and Mitigating Numerical Sources of Nondeterminism in LLM Inference](https://mlsanity.com/paper/2506.09501/) (2025)
- [Towards Robust Speech Representation Learning for Thousands of Languages](https://mlsanity.com/paper/2407.00837/) (2024)
- [Recent Advances in Discrete Speech Tokens: A Review](https://mlsanity.com/paper/2502.06490/) (2025)
- [Discrete Audio Tokens: More Than a Survey!](https://mlsanity.com/paper/2506.10274/) (2025)
- [ESPnet-Codec: Comprehensive Training and Evaluation of Neural Codecs for Audio, Music, and Speech](https://mlsanity.com/paper/2409.15897/) (2024)
- [Is Flash Attention Stable?](https://mlsanity.com/paper/2405.02803/) (2024)
- [UALM: Unified Audio Language Model for Understanding, Generation and Reasoning](https://mlsanity.com/paper/2510.12000/) (2025)
- [OpusLM: A Family of Open Unified Speech Language Models](https://mlsanity.com/paper/2506.17611/) (2025)
- [Optimizing Conversational Quality in Spoken Dialogue Systems with Reinforcement Learning from AI Feedback](https://mlsanity.com/paper/2601.19063/) (2026)
- [Bagpiper: Solving Open-Ended Audio Tasks via Rich Captions](https://mlsanity.com/paper/2602.05220/) (2026)
- [BSCodec: A Band-Split Neural Codec for High-Quality Universal Audio Reconstruction](https://mlsanity.com/paper/2511.06150/) (2025)

## Abstract (arXiv)

While Large Multimodal Models excel in comprehension, high-throughput inference engines lack native support for multimodal generation. This is severe in Speech Language Models, where generating multi-layered audio tokens via decoupled AR+NAR or synchronous Multi-Token Prediction (MTP) with delay-pattern interleaving conflicts with standard single-stream loops. We present a vLLM-based inference pipeline for unified speech understanding and generation. We extend autoregressive decoding to natively execute delay-pattern de-interleaving and coordinated multi-stream sampling, integrating an on-GPU acoustic decoder for end-to-end waveform synthesis. Crucially, we overcome the shared intuition that Classifier-Free Guidance (CFG) halves throughput. By co-scheduling paired conditional and unconditional requests within a continuous batch, our CFG implementation sustains 80% of non-CFG throughput, absorbing dual-request and logit merging overheads. We open-source our framework.