# Translate short sentence
**URL:**
**Author:**  [@Methos02](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FMethos02)%5C
**Post date:** [March 18, 2025, 2:20pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ftranslate-short-sentence%2F146293%2F1 "2025-03-18T14:20:04Z")
Hello everyone,
I am new to this field. I have set up a small API to dynamically translate my website. I have used several multilingual models because I need French, English, German, and Dutch (and soon Italian).
I am a beginner in Python, so I used ChatGPT to help me create this API.
Unfortunately, the translation quality is not very good…
- “Prénom” → “Name of family” in English
- “Envoyer” → “Send” in German
ChatGPT told me that the models I used were not designed for short sentences and that they work better with longer texts. It recommended using more traditional methods for this type of translation. I wanted to get your opinion and see which model you would recommend.
For now, I am leaning towards using a standard Google API for the site’s translation files and my API for translating user product descriptions.
Thanks in advance,
Methos
---
**Author:**  [@John6666](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FJohn6666)%5C
**Post date:** [March 18, 2025, 4:36pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ftranslate-short-sentence%2F146293%2F2 "2025-03-18T16:36:35Z")
That’s true. With short sentences and words, there’s not much benefit to using LLM. There are ways to use smaller LMs, etc., but it’s true that you can use Google Translate.
> **[Translation](https://huggingface.co/docs/transformers/tasks/translation)**
>
> We’re on a journey to advance and democratize artificial intelligence through open source and open science.
> **[OPUS-MT multilingual (TC+Bible) - a Helsinki-NLP Collection](https://huggingface.co/collections/Helsinki-NLP/opus-mt-multilingual-tcbible-670570b993cf2ffd1749334b)**
>
> multilingual translation models trained on the Tatoeba Translation Challenge dataset (from OPUS) and a massively multilingual Bible corpus
> [@Multilingual T5 Model Not Found?](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fmultilingual-t5-model-not-found%2F1892)%3A
>
> I could not able to see T5 Multilingual Model in Multilingual Model page of [Hugging Face](https://huggingface.co/transformers/multilingual.html) page but i can able to see Multilingual Model of T5 in [Google research](https://github.com/google-research/multilingual-t5) Page When can i able to see the model available in Hugging Face repo
> **[facebook/mbart-large-50 · Hugging Face](https://huggingface.co/facebook/mbart-large-50)**
>
> We’re on a journey to advance and democratize artificial intelligence through open source and open science.
* * *
To address the translation issues with short sentences on your website, here are some recommendations that balance ease of implementation with effectiveness:
1. **HuggingFace OPUS-MT Model** : Utilize the “Helsinki-NLP/opus-mt-mul-en” model, which is designed for multilingual translations. It’s user-friendly with the Transformers library and can handle multiple languages, including French, German, Dutch, and Italian.
2. **Facebook’s mBART or M2M Models** : Use the EasyNMT package to leverage these models, which are multilingual and support a wide range of languages. They are efficient for short sentences and can be integrated with minimal code.
3. **MarianMT Models** : Consider these for their speed and efficiency. They are optimized for real-time translations and come in various language pairs, suitable for quick processing on a website.
4. **DeepL API** : If budget allows, DeepL’s API offers high-quality translations, particularly effective for short texts, though it may have usage limits.
5. **Testing and Evaluation** : Develop a script to test different models with your specific short phrases. This hands-on approach will help you evaluate which model provides the best accuracy for your needs.
6. **Adjustments for Short Texts** : Ensure proper tokenization and padding are applied, as some models may require fixed input lengths. This adjustments can improve translation quality for short sentences.
**Implementation Steps** :
- Start with installing necessary libraries like `transformers` or `easynmt`.
- Use HuggingFace’s pipeline for OPUS-MT or EasyNMT for Facebook’s models.
- Consider setting up a test environment to evaluate each model’s performance.
By exploring these options, you can enhance the accuracy of your translations while maintaining ease of use, especially since you’re a Python beginner.
---
**Author:**  [@Methos02](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FMethos02)%5C
**Post date:** [March 20, 2025, 2:50pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ftranslate-short-sentence%2F146293%2F3 "2025-03-20T14:50:48Z")
Thank you very much, I’ll try that tomorrow afternoon!
---
**Author:**  [@Methos02](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FMethos02)%5C
**Post date:** [March 23, 2025, 11:47am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ftranslate-short-sentence%2F146293%2F4 "2025-03-23T11:47:27Z")
I tried reading your post, but it’s a bit complicated ^^ Is it really for beginners? xD
Here’s the code I have so far.
```auto
from fastapi import FastAPI
from pydantic import BaseModel
from transformers import pipeline
app = FastAPI()
pipe = pipeline(task='text2text-generation', model='facebook/m2m100_418M')
class TranslationRequest(BaseModel):
source: str
target: str
texts: dict
@app.post("/translate")
def translate(request: TranslationRequest):
translations = {}
prefix = "translate French to German: "
for key, text in request.texts.items():
original_text = text.strip()
result = pipe(
prefix + original_text,
forced_bos_token_id=pipe.tokenizer.get_lang_id(request.target)
)
translations[key] = result[0]["generated_text"]
return translations
```
I already have the code for Google Translate, but that means I would have to use both technologies since I’d have both short texts and slightly longer ones. Ideally, it would be best to have just one technology for both situations.
---
**Author:**  [@John6666](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FJohn6666)%5C
**Post date:** [March 23, 2025, 2:48pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ftranslate-short-sentence%2F146293%2F5 "2025-03-23T14:48:48Z")
Well, when it comes to creating APIs for the web, even if it’s aimed at beginners, it can be a little difficult…😅
---
**Author:**  [@Methos02](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FMethos02)%5C
**Post date:** [March 24, 2025, 9:37am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ftranslate-short-sentence%2F146293%2F6 "2025-03-24T09:37:03Z")
The API works, only the translation are bad ^^
---
**Author:**  [@John6666](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FJohn6666)%5C
**Post date:** [March 24, 2025, 10:06am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ftranslate-short-sentence%2F146293%2F7 "2025-03-24T10:06:16Z")
I think it’s often difficult to get satisfactory performance with the base model as it is. It’s a good idea to try out various models and, if they don’t seem to work, either fine-tune them or use a ready-made service.
[https://huggingface.co/models?pipeline\_tag=text2text-generation&library=transformers&sort=trending](https://huggingface.co/models?pipeline_tag=text2text-generation&library=transformers&sort=trending)
[https://huggingface.co/models?pipeline\_tag=translation&library=transformers&sort=trending](https://huggingface.co/models?pipeline_tag=translation&library=transformers&sort=trending)
---
**Author:**  [@Methos02](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FMethos02)%5C
**Post date:** [March 28, 2025, 1:49pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Ftranslate-short-sentence%2F146293%2F8 "2025-03-28T13:49:26Z")
Thx for your answer, I will look ASAP ! I have some bugs to fixe on the main branch ^^’