Build guide

How to build a translation service with Vikasit AI

An LLM translator handles many language pairs with context awareness — it understands idiom and tone better than rule-based systems. Add a glossary and tone instructions to keep terminology consistent.

Recommended model

Vikasit 3

Solid multilingual quality at a low price for general translation. Use Vikasit 3 Max for nuanced, literary, or domain-specific text.

Steps

1
Detect or accept the source and target languages.
2
Build a system prompt specifying target language, tone, and any glossary terms.
3
Pass the source text and request only the translation.
4
For long documents, translate paragraph by paragraph to preserve structure.
5
Optionally run a second pass to review fluency and terminology.
6
Cache common phrases to reduce repeat translation cost.

Code

The Vikasit Inference API is OpenAI-compatible, so this uses the standard OpenAI Python SDK pointed at https://api.vikasit.ai/v1.

translation.py

from openai import OpenAI

client = OpenAI(
    base_url="https://api.vikasit.ai/v1",
    api_key="sk-vikasit-...",  # get one at vikasit.ai/auth
)

def translate(text: str, target: str = "Hindi") -> str:
    resp = client.chat.completions.create(
        model="vikasit-3",
        messages=[
            {
                "role": "system",
                "content": f"Translate to {target}. Preserve tone. Output only the translation.",
            },
            {"role": "user", "content": text},
        ],
    )
    return resp.choices[0].message.content

Build your translation service today

Get an API key and 2M free tokens a day on Vikasit Nova. Pay-as-you-go, no minimums, OpenAI-compatible.

Get an API key See all models & pricing