CtrlK
BlogDocsLog inGet started
Tessl Logo

tessl/pypi-aleph-alpha-client

Python client to interact with Aleph Alpha API endpoints

Pending

Quality

Pending

Does it follow best practices?

Impact

Pending

No eval scenarios have been run

Overview
Eval results
Files

translation.mddocs/

Translation

Translate text between languages using Aleph Alpha's translation models. Provides high-quality translations with quality scoring and segment-level analysis.

Capabilities

Translation Requests

Configure translation operations with source text and target language specification.

class TranslationRequest:
    model: str
    source: str
    target_language: str
    
    def __init__(self, model: str, source: str, target_language: str):
        """
        Request for translation.
        
        Parameters:
        - model: The name of the model to be used for the translation
        - source: The input text to be translated
        - target_language: The desired target language using ISO 639 language codes
                         (e.g., "en" for English, "de" for German, "fr" for French)
        """

    def to_json(self) -> dict:
        """Convert the request to a JSON-serializable dictionary."""

Translation Responses

Structured response containing translated text with quality metrics and optional segment analysis.

class TranslationResponse:
    translation: str
    score: float
    segments: Optional[List[TranslationSegment]]
    num_tokens_prompt_total: int
    num_tokens_generated: int
    
    def __init__(
        self,
        translation: str,
        score: float,
        segments: Optional[List[TranslationSegment]],
        num_tokens_prompt_total: int,
        num_tokens_generated: int
    ):
        """
        Response from a translation request.
        
        Attributes:
        - translation: The complete translated output text
        - score: Overall quality estimate on a scale of 0 to 1
        - segments: List of translated segments (may be None)
        - num_tokens_prompt_total: Total tokens in prompt (may be zero)
        - num_tokens_generated: Total tokens generated (may be zero)
        """

    @classmethod
    def from_json(cls, json: dict) -> TranslationResponse:
        """Create a TranslationResponse from a JSON dictionary."""

Translation Segments

Individual segments of translated text with quality scoring.

class TranslationSegment:
    source: str
    translation: str
    score: float
    
    def __init__(self, source: str, translation: str, score: float):
        """
        A segment of translated text with its source and quality score.
        
        Parameters:
        - source: The input text to be translated
        - translation: The translated output text of the segment
        - score: Quality estimate for this segment on a scale of 0 to 1
        """

Translation Methods

Translate text using synchronous and asynchronous clients.

def translate(self, request: TranslationRequest) -> TranslationResponse:
    """
    Translate text to target language.
    
    Parameters:
    - request: Translation configuration
    
    Returns:
    TranslationResponse with translated text and quality metrics
    """

async def translate(self, request: TranslationRequest) -> TranslationResponse:
    """
    Translate text to target language (async).
    
    Parameters:
    - request: Translation configuration
    
    Returns:
    TranslationResponse with translated text and quality metrics
    """

Usage Examples

Basic translation with quality assessment:

from aleph_alpha_client import Client, TranslationRequest

client = Client(token="your-api-token")

# Simple translation
request = TranslationRequest(
    model="luminous-extended",
    source="Hello, how are you today?",
    target_language="de"  # German
)

response = client.translate(request)
print(f"Translation: {response.translation}")
print(f"Quality Score: {response.score:.2f}")

# Translation with segment analysis
request = TranslationRequest(
    model="luminous-extended", 
    source="Good morning! I hope you're having a wonderful day. The weather is beautiful today.",
    target_language="fr"  # French
)

response = client.translate(request)
print(f"Full Translation: {response.translation}")
print(f"Overall Quality: {response.score:.2f}")

if response.segments:
    print("\nSegment Analysis:")
    for i, segment in enumerate(response.segments):
        print(f"Segment {i+1}:")
        print(f"  Source: {segment.source}")
        print(f"  Translation: {segment.translation}")
        print(f"  Quality: {segment.score:.2f}")

# Async translation
import asyncio

async def translate_async():
    async with AsyncClient(token="your-api-token") as client:
        request = TranslationRequest(
            model="luminous-extended",
            source="Artificial intelligence is transforming our world.",
            target_language="es"  # Spanish
        )
        
        response = await client.translate(request)
        print(f"Async Translation: {response.translation}")
        return response

# Run async translation
response = asyncio.run(translate_async())

# Multiple language translation
languages = {
    "de": "German",
    "fr": "French", 
    "es": "Spanish",
    "it": "Italian"
}

source_text = "Technology connects people across the globe."

for lang_code, lang_name in languages.items():
    request = TranslationRequest(
        model="luminous-extended",
        source=source_text,
        target_language=lang_code
    )
    
    response = client.translate(request)
    print(f"{lang_name}: {response.translation} (Quality: {response.score:.2f})")

Install with Tessl CLI

npx tessl i tessl/pypi-aleph-alpha-client

docs

chat-interface.md

client-management.md

document-prompt-template.md

embeddings.md

evaluation.md

explanations.md

index.md

prompt-construction.md

steering.md

structured-output.md

text-completion.md

tokenization.md

translation.md

utilities.md

tile.json