Model reference · open weights
Ming-flash-omni-2.0 is an open-weight language model from inclusionAI. Ming-flash-omni-2.0 (BF16) weighs 209 GB; the smallest configuration that runs it is 2× H200 141 GB.
For self-hosting, compare this build's requirements below with the GPU you plan to rent.
Summary of the inclusionAI/Ming-flash-omni-2.0 model card, 2026-10-01
What it is
| Released by | inclusionAI |
|---|---|
| Released | 2026-02-10 |
| Parameters | 104.2B |
| VRAM | 209 GB for the weights |
What it runs on
| Card | Requests at once | Context max | Memory | |
|---|---|---|---|---|
| 8K each | 32K each | |||
| RTX 3060 12 GB … B200 180 GB 12 smaller cards | ||||
| 2× H200 141 GB tensor parallel | 89 | 22 | all 32K | 138 GB a card |
| 4× H100 80 GB tensor parallel | 120 | 30 | all 32K | 78.1 GB a card |
| 4× A100 80 GB tensor parallel | 172 | 43 | all 32K | 78.2 GB a card |
| 2× B200 180 GB tensor parallel | 216 | 54 | all 32K | 176 GB a card |
| 4× RTX PRO 6000 Blackwell 96 GB tensor parallel | 237 | 59 | all 32K | 93.8 GB a card |
| Requests at once | 8K tokens each | 32K tokens each |
|---|---|---|
| 1 | 212 GB | 214 GB |
| 5 | 214 GB | 222 GB |
| 8 | 216 GB | 229 GB |
| 16 | 220 GB | 246 GB |
| 32 | 229 GB | 280 GB |
| 64 | 246 GB | 349 GB |
One card, with vLLM's small-card settings.
From the model card
The newly released Ming-flash-omni 2.0 leverages the Ling-2.0 architecture—a Mixture-of-Experts (MoE) framework comprising 100B total and 6B active parameters. Representing a generational advancement over its predecessor, it establishes new State-of-the-Art (SOTA) benchmarks among open-source omni-MLLMs. Ming-flash-omni 2.0 effectively synergizes foundational abilities with specialized domain expertise. In particular, it exhibits superior performance in visual encyclopedic knowledge, immersive speech synthesis, and high-dynamic image generation and manipulation.
Compared to Ming-flash-omni Preview, Ming-flash-omni 2.0 focuses on optimizing capabilities across the following key domains:
Expert-level Multimodal Cognition: It accurately identifies plants and animals, recognizing cultural references (from regional cuisines to global landmarks), and delivering expert-level analysis of artifacts, including era, form, and craftsmanship. By synergizing high-resolution visual capture with a vast knowledge graph, the model achieves "vision-to-knowledge" synthesis, enabling superior knowledge understanding.
Immersive and Controllable Unified Acoustic Synthesis: Ming-flash-omni 2.0 introduces a unified end-to-end acoustic generation pipeline that integrates Speech, Audio, and Music within a single channel. Leveraging Continuous Autoregression coupled with a Diffusion Transformer (DiT) head, the model enables zero-shot voice cloning and nuanced attribute control (e.g., emotion, timbre, and ambient atmosphere). This architecture facilitates a transition from simple text-to-speech to highly expressive, emotionally resonant, and immersive auditory experiences.
High-Dynamic Controllable Image Generation and Manipulation: Ming-flash-omni 2.0 features a native multi-task architecture that unifies segmentation, generation, and editing, allowing for sophisticated spatiotemporal semantic decoupling. It excels in high-dynamic content creation, including atmospheric reconstruction, seamless scene composition, and context-aware object removal. By maintaining texture coherence and spatial depth consistency, Ming-flash-omni 2.0 achieves state-of-the-art precision in complex image manipulation tasks.
Enhanced Multimodal Cognition & Free Modality Switching
Streaming Video Conversation
Audio Context ASR & Dialect ASR
Controllable Image Generation
You can download our latest model from both Huggingface and ModelScope. For previous version model like Ming-flash-omni-Preview, Please refer to this link.
| Model | Input modality | Output modality | Download |
|---|---|---|---|
| Ming-flash-omni 2.0 | Image,text,video,audio | Image,text,audio | 🤗 HuggingFace 🤖 ModelScope |
| If you're in mainland China, we strongly recommend you to download our model from 🤖 ModelScope. |
pip install modelscope
modelscope download --model inclusionAI/Ming-flash-omni-2.0 --local_dir inclusionAI/Ming-flash-omni-2.0 --revision master
Note: This download process will take several minutes to several hours, depending on your network conditions.
pip install -r requirements.txt
pip install nvidia-cublas-cu12==12.4.5.8 # for H20 GPU
We provide a step-by-step running example:
Step 1 - Download the source code
git clone https://github.com/inclusionAI/Ming.git
cd Ming
Step 2 - Download the model weights and create a soft link to the source code directory
Download our model following Model Downloads
mkdir inclusionAI
ln -s /path/to/inclusionAI/Ming-flash-omni-2.0 inclusionAI/Ming-flash-omni-2.0
Step 3 - Enter the code directory, you can refer to the following codes to run the Ming-flash-omni model.
jupyter notebook cookbook.ipynb
We also provide a simple example on the usage of this repo. For detailed usage, please refer to [cookbook.ipynb](https://github.com/inclusionAI/Ming/bl
Quoted from the model card on Hugging Face. The full card is behind the Hugging Face link above.