Python AI API quickstart (₹ pricing, India)

Python + unoblox: text generation in rupees

Python developers already know the OpenAI SDK. Point it at unoblox, pick a model, and call chat completions billed in rupees on a monthly GST invoice from India, with input tax credit.

Minimal example (3 minutes)

pip install openai

from openai import OpenAI

client = OpenAI(
    base_url="https://api.unoblox.ai/v1",
    api_key="ub-gw-YOUR_API_KEY"  # from unoblox.ai/sign-in
)

response = client.chat.completions.create(
    model="qwen/qwen3-235b-a22b-instruct-2507",
    messages=[
        {"role": "user", "content": "Summarize GST in India"}
    ]
)

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

Run it, then check the request and its ₹ cost at /dashboard any time.

Advanced: streaming, vision, and async

# Streaming (real-time tokens)
stream = client.chat.completions.create(
    model="qwen/qwen3-235b-a22b-instruct-2507",
    messages=[{"role": "user", "content": "..."}],
    stream=True
)
for chunk in stream:
    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)

# Vision (if the model supports it)
response = client.chat.completions.create(
    model="qwen/qwen3-235b-a22b-instruct-2507",
    messages=[
        {
            "role": "user",
            "content": [
                {"type": "text", "text": "What's in this image?"},
                {"type": "image_url", "image_url": {"url": "https://..."}}
            ]
        }
    ]
)

# Async, for concurrent requests
import asyncio
from openai import AsyncOpenAI

aclient = AsyncOpenAI(base_url="https://api.unoblox.ai/v1", api_key="ub-gw-...")

async def ask(prompt):
    r = await aclient.chat.completions.create(
        model="deepseek-ai/deepseek-v4-flash",
        messages=[{"role": "user", "content": prompt}]
    )
    return r.choices[0].message.content

results = asyncio.run(asyncio.gather(*(ask(q) for q in ["q1", "q2", "q3"])))

Pricing: straightforward ₹ per token consumed

Model	Input ₹/1M	Output ₹/1M	Use case
Qwen3.8-27B	₹16.32	₹48.96	general, vision
DeepSeek V3.2	₹26.21	₹38.30	reasoning, coding
Qwen3 1.7B	₹0	₹0	free, rate-limited
Claude Sonnet	₹201.6	₹1008	production benchmark

Deploying behind Django or FastAPI

FastAPI's async views pair naturally with AsyncOpenAI:

from fastapi import FastAPI
from openai import AsyncOpenAI

app = FastAPI()
client = AsyncOpenAI(base_url="https://api.unoblox.ai/v1", api_key="ub-gw-...")

@app.post("/chat")
async def chat(prompt: str):
    r = await client.chat.completions.create(
        model="deepseek-ai/deepseek-v4-flash",
        messages=[{"role": "user", "content": prompt}]
    )
    return {"reply": r.choices[0].message.content}

Django is typically sync, so use the plain OpenAI client inside a view or a Celery task rather than AsyncOpenAI — Django's async view support works too if your project already uses it.

FAQ

Can I use asyncio for concurrent requests? Yes — see the async example above. Use AsyncOpenAI instead of OpenAI, then await calls inside async functions for non-blocking I/O.

How do I handle errors (timeouts, invalid key)? Catch openai.APIError and its subclasses (RateLimitError, AuthenticationError, and so on) — the same patterns as any OpenAI-compatible integration.

Do I need to install anything besides openai? No. The openai package alone is enough, on Python 3.8+.

Can I log every request and response? Yes. Attach a handler to Python's logging module and point the SDK's HTTP client logger at it, or wrap create() calls in your own logging decorator.

What's the difference between input and output tokens? Input is your prompt; output is the model's reply. Both are billed, usually at different rates, and unoblox itemizes both separately per request.

Can I batch multiple requests? unoblox doesn't offer an OpenAI-style batch endpoint yet. Loop individual requests instead — each is billed at the normal per-token rate.

Get started in rupees → https://unoblox.ai/sign-in