← all releases
vllm-project·Jan 31, 2026·v0.14.0·open source
vLLM-Omni
vLLM-Omni is an open-source framework that extends vLLM's support to omni-modality model inference and serving, encompassing text, image, video, and audio data processing. It introduces support for non-autoregressive architectures like Diffusion Transformers and offers heterogeneous outputs, facilitating complex model workflows.
CVLLMLLMs
overview
vLLM-Omni is an open-source framework that extends vLLM's support to omni-modality model inference and serving, encompassing text, image, video, and audio data processing. It introduces support for non-autoregressive architectures like Diffusion Transformers and offers heterogeneous outputs, facilitating complex model workflows.
key features
- 01Omni-modality support for text, image, video, and audio data processing
- 02Integration of non-autoregressive architectures such as Diffusion Transformers
- 03Heterogeneous outputs from traditional text generation to multimodal outputs
- 04Efficient KV cache management for state-of-the-art autoregressive support
- 05Pipelined stage execution overlapping for high throughput performance
related productsbrowse all →
Protenixopen sourceProtenix is an open-source, trainable PyTorch implementation of AlphaFold 3, designed for high-accuracy biomolecular structure prediction. It aims to advance accessible and extensible research tools for the computational biology community. ([github.com](https://github.com/bytedance/Protenix?utm_source=openai))Z-Imageopen sourceZ-Image is an efficient image generation foundation model developed by Tongyi-MAI, designed to produce high-quality, diverse, and stylistically versatile images. It serves as a robust base for creators, researchers, and developers seeking advanced image generation capabilities.Step-Audio-R1open sourceStep-Audio-R1 is an advanced audio language model developed by StepFun AI, designed to enhance audio reasoning capabilities by grounding its reasoning in acoustic features. It introduces Modality-Grounded Reasoning Distillation (MGRD), an iterative training framework that shifts the model's reasoning from textual abstractions to acoustic properties, effectively addressing the 'inverted scaling' problem where performance degrades with longer reasoning. This model has demonstrated superior performance across various audio understanding and reasoning benchmarks, surpassing models like Gemini 2.5 Pro and achieving results comparable to Gemini 3 Pro.