Metraipolis › archive › dc_6ea808a588a8
vLLM-Omni Technical Report: A Unified Serving Runtime for Omni-Modality Generation — arXiv:2610.09307
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.09307
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T11:28:58.791Z
- SHA-256 of the archived text
f19043786e3e098b571fe7edac819fe79ca9ebe8c7c68934e17bf6495ab627ff- Stable URL of this page
https://metraipolis.com/documento/dc_6ea808a588a8
How to cite
vLLM-Omni Technical Report: A Unified Serving Runtime for Omni-Modality Generation — arXiv:2610.09307 — arXiv (en), https://arxiv.org/abs/2610.09307 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6ea808a588a8 (Metraipolis archive dc_6ea808a588a8, sha256 f19043786e3e098b571fe7edac819fe79ca9ebe8c7c68934e17bf6495ab627ff)
BibTeX
@misc{dc6ea808a588a8,
title = {vLLM-Omni Technical Report: A Unified Serving Runtime for Omni-Modality Generation — arXiv:2610.09307},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.09307},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6ea808a588a8; sha256 f19043786e3e098b571fe7edac819fe79ca9ebe8c7c68934e17bf6495ab627ff},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.09307. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Interaction with intelligent systems is expanding beyond text-centric chatbots and coding agents. Speech-native assistants, visual generation and editing, world-model environments, and robot action loops require models that emit text, audio, images, video, and actions. These models differ in execution pattern: multi-stage autoregressive omni and TTS pipelines, iterative diffusion or flow-matching generators, and longer-lived world-model or robot loops that carry state across steps. As a result, serving is no longer a single text decode loop, but a heterogeneous multi-stage workflow with cross-stage transfer, streaming, and session-shaped interaction. Existing inference stacks are typically optimized for one architecture family. LLM servers deepen autoregressive scheduling and KV management, while diffusion stacks deepen denoising and parallel generation. Neither provides a shared control plane for pipelines that emit speech, pixels, or actions through separate generators, so production deployments often fall back to ad-hoc composition across disjoint runtimes. We present vLLM-Omni, a unified serving runtime for omni-modality generation. vLLM-Omni organizes each workload as a multi-stage pipeline under a single orchestrator that admits requests, advances them across stages, and demultiplexes streaming outputs. Specialized engines and stage replicas provide compute; a connector carries heavy payloads on the data plane; and session-oriented control supports long-lived duplex, world-model, and robot workloads. This report covers the architecture (stage-level KV paths, replica pools, multi-hardware platforms, and efficiency stack) and OpenAI-compatible and OpenPI APIs for omni, TTS, image/video, world-model, robot, and duplex workloads. We evaluate on the multimodal nightly CI on H100 (TTS and MiniCPM-o on H200), focused on Qwen3-Omni. Tema degli esploratori: calcolo e infrastrutture (elenco arXiv cs.DC, nuovi lavori del giorno). Categorie arXiv: Distributed, Parallel, and Cluster Computing (cs.DC). Pagina del lavoro: https://arxiv.org/abs/2610.09307 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-08T10:29:06.597Z dalla pagina https://arxiv.org/list/cs.DC/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).