---
license: apache-2.0
library_name: transformers
---

<div align='center'>
<h1>Emu3: Next-Token Prediction is All You Need</h1h1>
<h3></h3>

[Emu3 Team, BAAI](https://www.baai.ac.cn/english.html)

| [Project Page](https://emu.baai.ac.cn) | [Paper](https://baai-solution.ks3-cn-beijing.ksyuncs.com/emu3/Emu3-tech-report.pdf?KSSAccessKeyId=AKLTgew6Kdg6RsK92QSfB2KLA&Expires=2591406552&Signature=6BvwfLVqvfww26Bhwvk3mG0FrL8%3D) | [🤗HF Models](https://huggingface.co/collections/BAAI/emu3-66f4e64f70850ff358a2e60f) | [github](https://github.com/baaivision/Emu3) |


</div>

<div align='center'>
<img src="https://github.com/baaivision/Emu3/blob/main/assets/arch.png?raw=True" class="interpolation-image" alt="arch." height="80%" width="70%" />
</div>

We introduce **Emu3**, a new suite of state-of-the-art multimodal models trained solely with **<i>next-token prediction</i>**! By tokenizing images, text, and videos into a discrete space, we train a single transformer from scratch on a mixture of multimodal sequences.

### Emu3 excels in both generation and perception
**Emu3** outperforms several well-established task-specific models in both generation and perception tasks, surpassing flagship open models such as SDXL, LLaVA-1.6 and OpenSora-1.2, while eliminating the need for diffusion or compositional architectures.

<div align='center'>
<img src="https://github.com/baaivision/Emu3/blob/main//assets/comparison.png?raw=True" class="interpolation-image" alt="comparison." height="80%" width="80%" />
</div>

### Highlights

- **Emu3** is capable of generating high-quality images following the text input, by simply predicting the next vision token. The model naturally supports flexible resolutions and styles.
- **Emu3** shows strong vision-language understanding capabilities to see the physical world and provides coherent text responses. Notably, this capability is achieved without depending on a CLIP and a pretrained LLM.
- **Emu3** simply generates a video causally by predicting the next token in a video sequence, unlike the video diffusion model as in Sora. With a video in context, Emu3 can also naturally extend the video and predict what will happen next. 



#### Quickstart

```python
from PIL import Image
from transformers import AutoTokenizer, AutoModel, AutoImageProcessor, AutoModelForCausalLM
from transformers.generation.configuration_utils import GenerationConfig
import torch

import sys
sys.path.append(PATH_TO_BAAI_Emu3-Chat_MODEL)
from processing_emu3 import Emu3Processor

# model path
EMU_HUB = "BAAI/Emu3-Chat"
VQ_HUB = "BAAI/Emu3-VisionTokenizer"

# prepare model and processor
model = AutoModelForCausalLM.from_pretrained(
    EMU_HUB,
    device_map="cuda:0",
    torch_dtype=torch.bfloat16,
    attn_implementation="flash_attention_2",
    trust_remote_code=True,
)

tokenizer = AutoTokenizer.from_pretrained(EMU_HUB, trust_remote_code=True)
image_processor = AutoImageProcessor.from_pretrained(VQ_HUB, trust_remote_code=True)
image_tokenizer = AutoModel.from_pretrained(VQ_HUB, device_map="cuda:0", trust_remote_code=True).eval()
processor = Emu3Processor(image_processor, image_tokenizer, tokenizer)

# prepare input
text = "Please describe the image"
image = Image.open("assets/demo.png")

inputs = processor(
    text=text,
    image=image,
    mode='U',
    padding_side="left",
    padding="longest",
    return_tensors="pt",
)

# prepare hyper parameters
GENERATION_CONFIG = GenerationConfig(pad_token_id=tokenizer.pad_token_id, bos_token_id=tokenizer.bos_token_id, eos_token_id=tokenizer.eos_token_id)

# generate
outputs = model.generate(
    inputs.input_ids.to("cuda:0"),
    GENERATION_CONFIG,
    max_new_tokens=320,
)

outputs = outputs[:, inputs.input_ids.shape[-1]:]
print(processor.batch_decode(outputs, skip_special_tokens=True)[0])
```