> ## Documentation Index
> Fetch the complete documentation index at: https://docs.gnani.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Indian languages

> Get replies in the Indian language you need, translate, and classify messages written in Indian languages.

Early access
**Gnani Evon v3.5** is now available as an API to early access registrants.

## When to use this

* Your users write in an Indian language and expect replies in the same language.
* You need to translate between English and an Indian language.
* You need to classify or extract details from messages written in an Indian language.

## Supported languages

The model reads and writes 11 languages:

| Language | Script |
| - | - |
| English | Latin |
| Hindi | Devanagari |
| Bengali | Bengali |
| Telugu | Telugu |
| Tamil | Tamil |
| Marathi | Devanagari |
| Gujarati | Gujarati |
| Kannada | Kannada |
| Malayalam | Malayalam |
| Odia | Odia |
| Punjabi | Gurmukhi |

The model works best with each language in its own script. Romanised text (an Indian language written in Latin letters, such as "aapka order kab aayega") and code-mixed text (such as Hinglish) are not in this list, so test them with your own messages before you rely on them.

***

## Set the reply language

Say in the `system` message which language to reply in, and in which script. Name the language explicitly; do not rely on the model matching the user's language.

```bash cURL highlight={7} theme={null}
curl https://api.vachana.ai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "X-API-Key-ID: $GNANI_API_KEY" \
  -d '{
    "model": "evon-v3.5",
    "messages": [
      {"role": "system", "content": "You are a support assistant for a mobile wallet. Always reply in Kannada, in Kannada script, even if the user writes in English. Keep replies to two sentences."},
      {"role": "user", "content": "How do I check my wallet balance?"}
    ],
    "reasoning_effort": "none"
  }'
```

```python Python (OpenAI SDK) theme={null}
import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.vachana.ai/v1",
    api_key=os.environ["GNANI_API_KEY"],
)

SYSTEM = (
    "You are a support assistant for a mobile wallet. "
    "Always reply in Kannada, in Kannada script, even if the user writes in English. "
    "Keep replies to two sentences."
)

completion = client.chat.completions.create(
    model="evon-v3.5",
    messages=[
        {"role": "system", "content": SYSTEM},
        {"role": "user", "content": "How do I check my wallet balance?"},
    ],
    reasoning_effort="none",
)
print(completion.choices[0].message.content)
```

```javascript JavaScript (OpenAI SDK) theme={null}
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.vachana.ai/v1",
  apiKey: process.env.GNANI_API_KEY,
});

const SYSTEM =
  "You are a support assistant for a mobile wallet. " +
  "Always reply in Kannada, in Kannada script, even if the user writes in English. " +
  "Keep replies to two sentences.";

const completion = await client.chat.completions.create({
  model: "evon-v3.5",
  messages: [
    { role: "system", content: SYSTEM },
    { role: "user", content: "How do I check my wallet balance?" },
  ],
  reasoning_effort: "none",
});
console.log(completion.choices[0].message.content);
```

```python Python (requests) theme={null}
import os
import requests

SYSTEM = (
    "You are a support assistant for a mobile wallet. "
    "Always reply in Kannada, in Kannada script, even if the user writes in English. "
    "Keep replies to two sentences."
)

response = requests.post(
    "https://api.vachana.ai/v1/chat/completions",
    headers={"X-API-Key-ID": os.environ["GNANI_API_KEY"]},
    json={
        "model": "evon-v3.5",
        "messages": [
            {"role": "system", "content": SYSTEM},
            {"role": "user", "content": "How do I check my wallet balance?"},
        ],
        "reasoning_effort": "none",
    },
    timeout=120,
)
response.raise_for_status()
print(response.json()["choices"][0]["message"]["content"])
```

The reply is in `choices[0].message.content`, in Kannada script.

***

## How it works

The model reads and writes text in tokens. It has no separate language setting: the language of the reply comes from your instructions and the conversation. A clear instruction in the `system` message is the most reliable way to fix it, because the `system` message is sent with every request.

### Patterns for the system message

| Goal | Add to the `system` message |
| - | - |
| One fixed language | `Always reply in Tamil, in Tamil script.` |
| Match the user | `Reply in the language the user writes in. If you cannot tell, reply in Hindi.` |
| Script for speech | `Write numbers and amounts in words, for example "पाँच सौ रुपये" rather than "₹500".` |
| Keep terms in English | `Keep product names and order numbers exactly as written.` |

### Transliteration tips

* **Say which script you want.** "Reply in Hindi" can come back in Devanagari or Latin letters. "Reply in Hindi, in Devanagari script" is clearer.
* **Check the script of the reply** in your code if it matters, for example before you send text to a text-to-speech voice for one language.
* **Keep identifiers unchanged.** Ask the model not to transliterate names, IDs and codes your system needs to match.

***

## Examples

> [accordion] Translate English to Tamil
>
> ```json Request body theme={null}
> {
>   "model": "evon-v3.5",
>   "reasoning_effort": "none",
>   "temperature": 0.2,
>   "messages": [
>     {"role": "system", "content": "Translate the user's text into Tamil, in Tamil script. Reply with the translation only."},
>     {"role": "user", "content": "Your order has shipped and will arrive on Friday."}
>   ]
> }
> ```

> [accordion] Classify a Hindi message, with JSON output
>
> ```json Request body wrap theme={null}
> {
>   "model": "evon-v3.5",
>   "reasoning_effort": "none",
>   "temperature": 0,
>   "response_format": {
>     "type": "json_schema",
>     "json_schema": {
>       "name": "intent",
>       "schema": {
>         "type": "object",
>         "properties": {
>           "intent": {"type": "string", "enum": ["balance_enquiry", "failed_payment", "card_block", "other"]},
>           "language": {"type": "string"}
>         },
>         "required": ["intent", "language"]
>       }
>     }
>   },
>   "messages": [
>     {"role": "system", "content": "Classify the customer's message. Reply with JSON only."},
>     {"role": "user", "content": "मेरा UPI पेमेंट फेल हो गया लेकिन पैसे कट गए।"}
>   ]
> }
> ```
>
> The reply's `content` is a JSON string such as `{"intent": "failed_payment", "language": "hi"}`. See [JSON output](https://docs.gnani.ai/api/LLM/structured-outputs.md).

> [accordion] Reply in Kannada to an English question
>
> ```json Request body theme={null}
> {
>   "model": "evon-v3.5",
>   "reasoning_effort": "none",
>   "messages": [
>     {"role": "system", "content": "You are a bank's assistant. Always reply in Kannada, in Kannada script, in one or two sentences."},
>     {"role": "user", "content": "What documents do I need to open a savings account?"}
>   ]
> }
> ```

> [accordion] Summarise a Hindi message in English
>
> ```json Request body theme={null}
> {
>   "model": "evon-v3.5",
>   "reasoning_effort": "none",
>   "messages": [
>     {"role": "system", "content": "Summarise the user's message in one English sentence for a support agent."},
>     {"role": "user", "content": "मैंने कल ₹2,500 का रिचार्ज किया था लेकिन अभी तक बैलेंस नहीं आया।"}
>   ]
> }
> ```

***

## Voice agents

When the reply is spoken by a text-to-speech voice, speed and length matter more than in a chat window.

* **Turn thinking off** with `"reasoning_effort": "none"`. Thinking adds time before the first word of the answer. See [Thinking](https://docs.gnani.ai/api/LLM/thinking.md).
* **Ask for short replies.** One or two sentences per turn, no lists or markdown. Set `max_tokens` to match.
* **Stream the reply** with `"stream": true`, and send text to speech as it arrives. See [Streaming](https://docs.gnani.ai/api/LLM/streaming.md).
* **Write for the ear.** Ask for numbers, amounts and dates in words, in the reply language.
* **Match the languages end to end.** Pick the speech-to-text language and the text-to-speech voice to match the reply language. See [Speech-to-Text models](https://docs.gnani.ai/api/STT/models/overview.md) and [Text-to-Speech](https://docs.gnani.ai/api/TTS/tts-inference.md).

***

## Gotchas

* **Indian scripts can use more tokens.** Text in Indian scripts can take more tokens than English for the same meaning, so the same conversation uses more of your context and daily quota. Every response reports the exact count in `usage`. See [Limits](https://docs.gnani.ai/api/LLM/limits.md).
* **The instruction must be in every request.** The API is stateless, so keep the language instruction in the `system` message you send each time.
* **Personal identifiers are masked in replies**, such as phone numbers and UPI IDs, so do not ask the model to repeat them back. See [Safety and privacy](https://docs.gnani.ai/api/LLM/safety.md).

## Related

> [card] [JSON output](https://docs.gnani.ai/api/LLM/structured-outputs.md)
>
> Get replies your code can parse.

> [card] [Limits](https://docs.gnani.ai/api/LLM/limits.md)
>
> Tokens, context size and the daily quota.

> [card] [Speech-to-Text models](https://docs.gnani.ai/api/STT/models/overview.md)
>
> Transcribe what the caller says.

> [card] [Text-to-Speech](https://docs.gnani.ai/api/TTS/tts-inference.md)
>
> Speak the model's reply.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.