GPT-5 nano API in India

GPT-5 nano is the smallest, lowest-cost entry point into the GPT-5 family, built for the kind of workload where you're making an enormous number of small calls rather than a few large ones. On unoblox, that same model is available in India through one ₹-billed, OpenAI-compatible endpoint, so the economics of running it at scale are visible in rupees from the first call.

What GPT-5 nano is actually for

Nano-tier models are not meant to compete with flagship reasoning quality — they're meant to be cheap and fast enough that you stop thinking about cost per call at all. That makes GPT-5 nano a fit for tagging, routing, short classification, moderation pre-filters, and any pipeline step that runs on every single request or every row of a dataset, where GPT-5 or even GPT-5 mini would be disproportionately expensive for the value each call adds.

₹ pricing for GPT-5 nano

GPT-5 nano isn't one of the models with a fixed indicative rate published in this note — its live ₹ rate is always visible on its model page before you send a single token. For scale, here's where the rest of the GPT-5 line sits:

Model	Input (₹ / 1M tokens)	Output (₹ / 1M tokens)
GPT-5 mini	25.2	201.6
GPT-5	126	1008

As the smallest tier, GPT-5 nano is priced below GPT-5 mini — check the model page for the exact current number rather than relying on any figure quoted elsewhere.

Calling GPT-5 nano from the unoblox endpoint

curl https://api.unoblox.ai/v1/chat/completions \
  -H "Authorization: Bearer ub-gw-xxxxxxxxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-5-nano",
    "messages": [{"role": "user", "content": "Classify this ticket as billing, technical, or sales."}]
  }'

Because every model on unoblox sits behind the same base URL and key, moving a pipeline step from GPT-5 nano up to GPT-5 mini (or back down) later is a one-line change.

Choosing between GPT-5 nano, mini and full GPT-5

• GPT-5 nano — highest volume, simplest task, cost matters most.
• GPT-5 mini — production features where quality needs to be a step above nano but full GPT-5 would be overkill.
• GPT-5 — the hardest subset of requests, where getting the answer right matters more than shaving cost per call.

A common pattern is to run everything through nano first as a cheap filter, and escalate only the requests that need it to mini or full GPT-5.

Frequently asked questions

Is GPT-5 nano available in India via unoblox? Yes, openai/gpt-5-nano is callable immediately on https://api.unoblox.ai/v1, billed in rupees.

What does GPT-5 nano cost on unoblox? It isn't in our fixed indicative price list; the live ₹ rate is shown on its model page and is lower than GPT-5 mini's.

What is GPT-5 nano best used for? High-volume, low-complexity steps — classification, routing, tagging, short extraction — where per-call cost needs to be as small as possible.

Does using GPT-5 nano mean my data is processed in India? No. It runs on OpenAI's infrastructure; the India-specific benefit is ₹ billing, a GST invoice and one endpoint, not data residency.

Can I mix GPT-5 nano with other models in the same application? Yes. The same ub-gw- key works across the whole catalogue, so a pipeline can call nano for filtering and a larger model for the final answer.

Is there a GST invoice for GPT-5 nano usage? Yes, usage across all models including GPT-5 nano is combined into one monthly GST invoice from an Indian entity, with input tax credit claimable.

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