---
title: m2m100_418M
canonical_url: "https://www.modelscope.cn/models/cubeai/m2m100_418M"
md_url: "https://www.modelscope.cn/models/cubeai/m2m100_418M.md"
repository: cubeai/m2m100_418M
chinese_name: "多语言翻译(m2m100_418M)"
last_updated: 2024-03-06
license: "Apache License 2.0"
pipeline_tag: translation
tasks:
  - translation
model_type:
  - m2m_100
architectures:
  - M2M100ForConditionalGeneration
library_name:
  - pytorch
frameworks:
  - Pytorch
downloads: 583
stars: 5
---

# m2m100_418M

> m2m100_418M - cubeai 在 ModelScope 开源的模型。M2M100 is a multilingual encoder-decoder (seq-to-seq) model trained for Many-to-Many multilingual translation. It was introduced in this paper and first released in this repository .

cubeai/m2m100_418M 是 ModelScope 魔搭社区上的translation模型，采用 Apache License 2.0 许可。

- **Repository**: cubeai/m2m100_418M
- **License**: Apache License 2.0
- **Tasks**: translation
- **Downloads**: 583
- **Stars**: 5
- **Last updated**: 2024-03-06

Source: https://www.modelscope.cn/models/cubeai/m2m100_418M

---

# 多语言翻译(m2m100_418M)

支持100种语言的相互翻译。

M2M100 is a multilingual encoder-decoder (seq-to-seq) model trained for Many-to-Many multilingual translation. It was introduced in [this paper](https://arxiv.org/abs/2010.11125) and first released in [this repository](https://github.com/pytorch/fairseq/tree/master/examples/m2m_100) .

The model that can directly translate between the 9,900 directions of 100 languages. To translate into a target language, the target language id is forced as the first generated token. To force the target language id as the first generated token, pass the forced_bos_token_id parameter to the generate method.

## 使用

https://openi.pcl.ac.cn/cubeai-model-zoo/hf_facebook_m2m100_418M

## 模型来源

https://hf-mirror.com/facebook/m2m100_418M