Tarjamli v3 Translation API

Translate Moroccan Arabic written in Arabic or Latin script into English, French, or Modern Standard Arabic. Tarjamli is a focused translation model. It does not provide explanations, chat, or analysis.

Endpoint

text

POST https://api.sawalni.com/v1/chat/completions

Model contract

Field	Value
Model	`tarjamli-v3`
Source	Moroccan Arabic in Arabic or Latin script
Targets	`English`, `French`, `Modern Standard Arabic`
Context window	8,192 tokens, including input and output
Output	The raw translation only

Use exactly this user-message format:

text

Translate this Moroccan Arabic text to TARGET. Return only the translation.

SOURCE_TEXT

Replace TARGET with one of the three target names in the table. Do not add a system message, chat history, labels, Markdown, or extra instructions. Use temperature: 0 for deterministic translation.

cURL

bash

curl https://api.sawalni.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "tarjamli-v3",
    "messages": [{
      "role": "user",
      "content": "Translate this Moroccan Arabic text to English. Return only the translation.\n\nواش نتا جاي غدا ولا باقي ما عارف؟"
    }],
    "temperature": 0,
    "max_tokens": 512,
    "stream": false
  }'

Python

python

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://api.sawalni.com/v1",
)

response = client.chat.completions.create(
    model="tarjamli-v3",
    messages=[{
        "role": "user",
        "content": (
            "Translate this Moroccan Arabic text to French. "
            "Return only the translation.\n\n"
            "safi ana ghadi nmchi daba, n3ayet lik mn b3d"
        ),
    }],
    temperature=0,
    max_tokens=512,
)

print(response.choices[0].message.content)

Response

Read only choices[0].message.content. That field contains the translation without a label or explanation.

json

{
  "model": "tarjamli-v3",
  "choices": [{
    "message": {
      "role": "assistant",
      "content": "Are you coming tomorrow or are you still unsure?"
    },
    "finish_reason": "stop"
  }]
}

Use the Tarjamli playground to verify a request before integration.

Model contract

Field	Value
Model	`tarjamli-v3`
Source	Moroccan Arabic in Arabic or Latin script
Targets	`English`, `French`, `Modern Standard Arabic`
Context window	8,192 tokens, including input and output
Output	The raw translation only

Use exactly this user-message format:

text

Translate this Moroccan Arabic text to TARGET. Return only the translation.

SOURCE_TEXT

cURL

bash

curl https://api.sawalni.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "tarjamli-v3",
    "messages": [{
      "role": "user",
      "content": "Translate this Moroccan Arabic text to English. Return only the translation.\n\nواش نتا جاي غدا ولا باقي ما عارف؟"
    }],
    "temperature": 0,
    "max_tokens": 512,
    "stream": false
  }'

Python

python

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://api.sawalni.com/v1",
)

response = client.chat.completions.create(
    model="tarjamli-v3",
    messages=[{
        "role": "user",
        "content": (
            "Translate this Moroccan Arabic text to French. "
            "Return only the translation.\n\n"
            "safi ana ghadi nmchi daba, n3ayet lik mn b3d"
        ),
    }],
    temperature=0,
    max_tokens=512,
)

print(response.choices[0].message.content)

Response

Read only choices[0].message.content. That field contains the translation without a label or explanation.

json

{
  "model": "tarjamli-v3",
  "choices": [{
    "message": {
      "role": "assistant",
      "content": "Are you coming tomorrow or are you still unsure?"
    },
    "finish_reason": "stop"
  }]
}

Use the Tarjamli playground to verify a request before integration.