> For clean Markdown of any page, append `.md` to the page URL.
> For a complete documentation index, see https://docs.sarvam.ai/llms.txt.
> For full documentation content in one file, see https://docs.sarvam.ai/llms-full.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.sarvam.ai/_mcp/server.

# How to control response randomness with `temperature`

> Control how focused or varied model responses are.

The `temperature` parameter controls how **focused or varied** model responses are.

**Range:** `0` to `2`\
**Default:** `0.2`

* **Lower temperature** → more focused, predictable answers (e.g. `0.2`)
* **Higher temperature** → more creative, varied responses (e.g. `0.8` or `1.0`)

👉 **Tip**: For most use cases, values between `0.2` and `0.8` give good results.

> **Note**
>
> Output is not bit-reproducible, even with `temperature=0`. Use a lower temperature to
> reduce variation, not to guarantee identical responses.

### How it works:

| **Mode**              | **Recommended `temperature`** | **Behavior**                        |
| --------------------- | ----------------------------- | ----------------------------------- |
| Non-thinking mode     | `0.2` *(default)*             | Straightforward, factual responses  |
| Thinking mode         | `0.5` or higher               | Deeper reasoning, more exploration  |
| Highly creative       | `0.8` - `1.0`                 | Storytelling, brainstorming, poetry |
| Very random / playful | `> 1.0`                       | Unexpected, experimental output     |

First, install the SDK:

```bash
pip install -Uqq sarvamai
```

Then use the following Python code:

```python
from sarvamai import SarvamAI

# Initialize the SarvamAI client with your API key
client = SarvamAI(api_subscription_key="YOUR_SARVAM_API_KEY")

# Example 1: Using default temperature (0.2): straightforward, factual response
response = client.chat.completions(
    model="sarvam-105b",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Explain the concept of gravity."}
    ],
    # temperature is not specified → uses default 0.2
)

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

```python
from sarvamai import SarvamAI

client = SarvamAI(api_subscription_key="YOUR_SARVAM_API_KEY")

# Example 2: Using temperature = 0.9: more creative, varied response
response = client.chat.completions(
    model="sarvam-105b",
    messages=[
        {"role": "system", "content": "You are a creative storyteller."},
        {"role": "user", "content": "Tell me a story about a magical tiger."}
    ],
    temperature=0.9  # More creative storytelling
)

# Receive assistant's reply as output
print(response.choices[0].message.content)
```

> Control how focused or varied model responses are.