Amazon Translate logo

Amazon Translate

Amazon Translate

UpFreeOpen Sourcecloudby Amazon Web Services62· JavaScript· MIT

Amazon Translate is the AWS neural machine translation service supporting 70+ language pairs for text and documents.

Visit site ↗Source ↗Health checked 9h ago
Use it when

70+ language pairs

Watch for

Translation quality lags DeepL and GPT-4 on many pairs

First check

Call Translate.translateText for short synchronous translation, or startTextTranslationJob for async batch processing of S3 documents.

Auth
api_key
CORS
No
HTTPS
Yes
Signup
Required
Latency
10 ms
Protocol
REST
Pricing
paid
Stars
62

Uptime · 30-day window

Probes: 1Uptime: 100%Avg latency: 10ms

GitHub activity

62JavaScriptMIT17 open issuesLast commit 110d ago
01

About this API

Translate is the AWS neural machine translation service covering 70+ language pairs. Quality is adequate on generic business text but lags DeepL, Google Translate, or GPT-4 head-to-head, especially on literary text or long contextual passages. The differentiator is AWS integration and customizability: Custom Terminology lets you upload glossaries (e.g. force "Stripe" to translate as "Stripe" instead of literally "stripe"); Active Custom Translation (ACT) lets you upload parallel bilingual data to fine-tune for your domain.

Two usage modes. translateText is sync — up to 5,000 characters per call, millisecond latency. startTextTranslationJob is async batch — translate an entire S3 folder of .txt / .html / .docx / .pptx / .xlsx documents and write results to another S3 path.

If you need high quality, tone preservation, or strong long-context handling, evaluate DeepL (excellent for European languages), Google Translate (widest language support), or call an LLM directly (GPT-4 / Claude). Amazon Translate is best for AWS-native, large-volume, cost-sensitive workloads where "good enough" suffices.

02

What you can build

  • 1Real-time translation of user-generated comments and posts
  • 2Localize product copy and multilingual content
  • 3Route customer support tickets across languages
  • 4Batch document translation
03

Strengths & limitations

Strengths

  • 70+ language pairs
  • Custom Terminology controls brand/technical term translation
  • Active Custom Translation fine-tunes on your bilingual data

Limitations

  • Translation quality lags DeepL and GPT-4 on many pairs
  • Some low-resource languages are basic-only
  • Short context window — long passages lose coherence
04

Example request

Generic template — replace <endpoint> with the real path from the docs.
curl https://github.com/mermade/aws2openapi/<endpoint> \
  -H "Authorization: Bearer $API_KEY"
# Some providers use X-Api-Key instead — verify in the docs.
05

Getting started

Call Translate.translateText for short synchronous translation, or startTextTranslationJob for async batch processing of S3 documents.

06

FAQ

What is the free tier?+

First 12 months: 2M characters/month free. Beyond that, ~$15 per million characters.

How does it compare to DeepL?+

DeepL usually leads on European languages; Translate is competitive with Google on Asian languages. Run your own benchmark for specific use cases.

How do I keep brand names from being translated?+

Upload a Custom Terminology CSV mapping source to target terms, then reference it via the TerminologyNames parameter at call time.

07

Technical details

CORS: NoHTTPS: YesSignup: YesOpen source: Yes
Auth type
api_key
Pricing
paid
Protocols
REST
SDKs
python, javascript, java, go, ruby, csharp
Response time
10 ms
Last health check
5/12/2026, 7:36:34 AM
08

Endpoints

Parsed from the OpenAPI spec. Showing 12 of 18 non-deprecated endpoints.

POST
/#X-Amz-Target=AWSShineFrontendService_20170701.CreateParallelData
X-Amz-Target:header*
POST
/#X-Amz-Target=AWSShineFrontendService_20170701.DeleteParallelData
X-Amz-Target:header*
POST
/#X-Amz-Target=AWSShineFrontendService_20170701.DeleteTerminology
X-Amz-Target:header*
POST
/#X-Amz-Target=AWSShineFrontendService_20170701.DescribeTextTranslationJob
X-Amz-Target:header*
POST
/#X-Amz-Target=AWSShineFrontendService_20170701.GetParallelData
X-Amz-Target:header*
POST
/#X-Amz-Target=AWSShineFrontendService_20170701.GetTerminology
X-Amz-Target:header*
POST
/#X-Amz-Target=AWSShineFrontendService_20170701.ImportTerminology
X-Amz-Target:header*
POST
/#X-Amz-Target=AWSShineFrontendService_20170701.ListLanguages
MaxResults:queryNextToken:queryX-Amz-Target:header*
POST
/#X-Amz-Target=AWSShineFrontendService_20170701.ListParallelData
MaxResults:queryNextToken:queryX-Amz-Target:header*
POST
/#X-Amz-Target=AWSShineFrontendService_20170701.ListTagsForResource
X-Amz-Target:header*
POST
/#X-Amz-Target=AWSShineFrontendService_20170701.ListTerminologies
MaxResults:queryNextToken:queryX-Amz-Target:header*
POST
/#X-Amz-Target=AWSShineFrontendService_20170701.ListTextTranslationJobs
MaxResults:queryNextToken:queryX-Amz-Target:header*

6 more endpoints not shown. See the OpenAPI spec for the full list.

09

Tags

10

More from Amazon Web Services