# RadixArk: DaisyRose 27B on AICredits

Guide for calling `radixark/daisyrose-27b` through the AICredits API. Paste it into a coding agent to add this model to a project.

## Model

- Model ID: `radixark/daisyrose-27b`
- Type: chat
- Context window: 262K tokens
- Input: text, image
- Output: text

## Pricing

- ₹14.63 per 1M input tokens
- ₹183 per 1M output tokens
- ₹3.66 per 1M cached input tokens

Prices are in Indian rupees and are deducted from your prepaid wallet. They follow the live USD/INR rate, so read the current figure from `GET https://api.aicredits.in/v1/models` or https://aicredits.in/models/radixark/daisyrose-27b.

## Authentication

Create an API key (it starts with `sk-`) in Dashboard → API Keys, then send it as a bearer token:

```text
Authorization: Bearer $AICREDITS_API_KEY
```

## API

### Chat Completions API (OpenAI format)

POST https://api.aicredits.in/v1/chat/completions

```python
import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["AICREDITS_API_KEY"],
    base_url="https://api.aicredits.in/v1",
)

response = client.chat.completions.create(
    model="radixark/daisyrose-27b",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)
```

```typescript
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.AICREDITS_API_KEY,
  baseURL: "https://api.aicredits.in/v1",
});

const response = await client.chat.completions.create({
  model: "radixark/daisyrose-27b",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);
```

```bash
curl https://api.aicredits.in/v1/chat/completions \
  -H "Authorization: Bearer $AICREDITS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "radixark/daisyrose-27b",
    "messages": [{ "role": "user", "content": "Hello!" }]
  }'
```

### Messages API (Anthropic format)

POST https://api.aicredits.in/v1/messages. The Anthropic SDK adds `/v1/messages` itself, so its base URL has no `/v1`.

```python
import os
import anthropic

client = anthropic.Anthropic(
    api_key=os.environ["AICREDITS_API_KEY"],
    base_url="https://api.aicredits.in",
)

message = client.messages.create(
    model="radixark/daisyrose-27b",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello!"}],
)
print(message.content)
```

### Responses API (OpenAI format)

POST https://api.aicredits.in/v1/responses

```bash
curl https://api.aicredits.in/v1/responses \
  -H "Authorization: Bearer $AICREDITS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "radixark/daisyrose-27b",
    "input": "Hello!"
  }'
```

## Errors

| Status | Message contains | What to do |
|---|---|---|
| 401 | `Invalid API Key` | The key is wrong, revoked or from another provider. Use an AICredits key (it starts with sk-) from Dashboard → API Keys. |
| 402 | `API Key Budget Exceeded` | This key has used its budget. Raise the budget in Dashboard → API Keys, or use another key. |
| 402 | `Insufficient Balance` | The wallet cannot cover the request. Top up in Dashboard → Billing. |
| 404 | `Unknown endpoint: POST /v1/v1/…` | The base URL is wrong, most often /v1 typed twice. OpenAI-format clients use https://api.aicredits.in/v1; Anthropic-format clients use https://api.aicredits.in. |
| 404 | `Model '…' is not available` | The model ID is not in the catalog. IDs are written provider/model; copy one from the models page or GET /v1/models. |

Provider errors, timeouts and empty replies are not billed. Which failed or cancelled calls are billed: https://aicredits.in/docs/troubleshooting/failed-calls-and-billing

## More

- Model page: https://aicredits.in/models/radixark/daisyrose-27b
- All docs: https://aicredits.in/docs
- Index for agents: https://aicredits.in/llms.txt
