ByteDance-Seed/Seed-X-7B

ByteDance-Seed/Seed-X-7B

Language: Python

License: NOASSERTION

Stars: 172

Forks: 7

Open issues: 15

Created: 2025-07-16T07:18:16Z

Pushed: 2025-08-18T11:40:20Z

Default branch: main

Fork: no

Archived: no

README:

You can get to know us better through the following channels👇

Seed-X: Building Strong Multilingual Translation LLM with 7B Parameters

We are excited to introduce Seed-X, a powerful series of open-source multilingual translation language models, including an instruction model, a reinforcement learning model, and a reward model. It pushes the boundaries of translation capabilities within 7 billion parameters.

📢 News

[2025/07/31] 🔥 We have uploaded a new version of the challenge set along with manually annotated reference translations.

[2025/07/28] 🔥 We have deployed our Seed-X-PPO on 🤗 HuggingFace Spaces. Welcome to try out our model!

[2025/07/28] 🔥 We have released quantized 8-bit and 4-bit PPO models.

[2025/07/18] 🔥 We have released the Seed-X-Challenge-Set.

[2025/07/18] 🔥 Seed-X-Instruct/PPO/RM are now avaliable on Huggingface!

🌟 Highlights

• Exceptional translation capabilities: Seed-X exhibits state-of-the-art translation capabilities, on par with or outperforming ultra-large models like Gemini-2.5, Claude-3.5, and GPT-4, as validated by human evaluations and automatic metrics.
• Deployment and inference-friendly: With a compact 7B parameter count and mistral architecture, Seed-X offers outstanding translation performance in a lightweight and efficient package, ideal for deployment and inference.
• Broad domain coverage: Seed-X excels on a highly challenging translation test set spanning diverse domains, including the internet, science and technology, office dialogues, e-commerce, biomedicine, finance, law, literature, and entertainment.

🏆 Performance

We evaluated Seed-X on a diverse set of translation benchmarks, including FLORES-200, WMT-25, and a publicly released challenge set accompanied by human evaluations.


For detailed benchmark results and analysis, please refer to our Technical Report.

⚡ Quick Start

We are excited to introduce Seed-X, featuring three powerful models:

| Model Name | Description | Download | | ----------- | ----------- |----------- | Seed-X-Instruct | Instruction-tuned for alignment with user intent. |🤗 Model| | Seed-X-PPO | RL trained to boost translation capabilities. | 🤗 Model| | Seed-X-PPO-GPTQ-Int8 | Quantization: GPTQ 8-bit. | 🤗 Model| | Seed-X-PPO-AWQ-Int4 | Quantization: AWQ 4-bit. | 🤗 Model| | Seed-X-RM | Reward model to evaluate the quality of translation.| 🤗 Model|

👉 Deploying Seed-X-PPO with ``vllm

❗The language tags at the end of the prompt is necessary, which are used in PPO training. For example, when the target language is German, \ needs to be added. You can refer to the above table for language abbreviations.

❗This model is specialized in multilingual translation, which is unexpected to support other tasks.

❗We don't have any chat template, thus you don't have to perform ``tokenizer.apply_chat_template``. Please avoid prompting the model in a multi-round conversation format.

❗We recommend against using unofficial quantized versions for local deployment. We will soon release an official quantized model and develop a demo on Hugging Face Space.

Here is a simple example demonstrating how to load the model and perform translation using ``vllm

Recommended: ``vllm==0.8.0, transformers==4.51.3

from vllm import LLM, SamplingParams, BeamSearchParams

model_path = "./ByteDance-Seed/Seed-X-PPO-7B"

model = LLM(model=model_path,
max_num_seqs=512,
tensor_parallel_size=8,
enable_prefix_caching=True,
gpu_memory_utilization=0.95)

messages = [
# without CoT
"Translate the following English sentence into Chinese:\nMay the force be with you ",
# with CoT
"Translate the following English sentence into Chinese and explain it in detail:\nMay the force be with you "
]

# Beam search (We recommend using beam search decoding)
decoding_params = BeamSearchParams(beam_width=4,
max_tokens=512)
# Greedy decoding
decoding_params = SamplingParams(temperature=0,
max_tokens=512,
skip_special_tokens=True)

results = model.generate(messages, decoding_params)
responses = [res.outputs[0].text.strip() for res in results]

print(responses)

License

This project is licensed under OpenMDW. See the LICENSE file for details.

Citation

@misc{cheng2025seedxbuildingstrongmultilingual,
title={Seed-X: Building Strong Multilingual Translation LLM with 7B Parameters},
author={Shanbo Cheng and Yu Bao and Qian Cao and Luyang Huang and Liyan Kang and Zhicheng Liu and Yu Lu and Wenhao Zhu and Jingwen Chen and Zhichao Huang and Tao Li and Yifu Li and Huiying Lin and Sitong Liu and Ningxin Peng and Shuaijie She and Lu Xu and Nuo Xu and Sen Yang and Runsheng Yu and Yiming Yu and Liehao Zou and Hang Li and Lu Lu and Yuxuan Wang and Yonghui Wu},
year={2025},
eprint={2507.13618},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2507.13618},
}

About ByteDance Seed Team

Founded in 2023, ByteDance Seed Team is dedicated to crafting the industry's most advanced AI foundation models. The team aspires to become a world-class research team and make significant contributions to the advancement of science and society.