GPT-4 pricing in India (₹)

GPT-4 pricing in rupees — no international payment friction

OpenAI's GPT models are now priced and billed directly in Indian rupees through unoblox. One OpenAI-compatible endpoint, one API key, monthly GST invoices — rupee billing only, transparent costs.

Real GPT-4 pricing (per 1M tokens)

Model	Input (₹)	Output (₹)	Use case
GPT-5	₹126	₹1,008	Complex reasoning, long-context
GPT-5 mini	₹25.2	₹201.6	Cost-optimized, most tasks
GPT-4o	₹252	₹1,008	Balanced, multimodal
GPT-4.1	₹201.6	₹806.4	Precise, reliable

Why rupee billing matters

Many Indian teams struggle with international payment methods. This creates friction:

• No international payment option available? Blocked from using GPT.
• Price fluctuations are unpredictable.
• GST tracking becomes manual and error-prone.

unoblox inverts this. Base URL stays https://api.unoblox.ai/v1, your API key is ub-gw-…, and billing happens on a clean monthly ₹ invoice. Transparent pricing, no payment friction, claimable GST input credit for registered businesses.

Quick comparison: GPT-5 vs GPT-5 mini

• GPT-5 (₹126/M input): for research, planning, multimodal reasoning. Overkill for simple categorization.
• GPT-5 mini (₹25.2/M input): 5× cheaper per token, 99% of the performance for most production workloads.

For a typical chatbot handling 10M monthly input tokens, GPT-5 costs ₹1,260; GPT-5 mini costs ₹252 — same outputs, five-fold savings.

How to switch

If you're already calling GPT, migration takes 2 lines:

from openai import OpenAI
client = OpenAI(api_key="ub-gw-YOUR_KEY", base_url="https://api.unoblox.ai/v1")

Streaming, tool use, JSON mode, and vision all work unchanged.

Frequently asked questions

Do I need an international payment method? No. unoblox accepts Indian payment methods (Razorpay, bank transfer) and invoices in rupees only.

Can I claim GST input credit on the invoice? Yes. The invoice itemizes the ₹ spend. Register with unoblox, and the monthly statement is ready for ITC claims.

Is this the official OpenAI API? unoblox routes through OpenAI's backend, so latency and quality are identical. We add the rupee layer on top.

What if GPT-4o is cheaper for my use case? Use the pricing table above to calculate: token count × model rate. Most chatbots save 20–40% switching from GPT-4.1 to GPT-4o.

How fast is token delivery? Same as OpenAI direct — we proxy transparently. Latency is <500ms for text, <3s for vision.

Can I use this with LangChain or Vercel AI SDK? Yes. Both auto-detect OpenAI compatibility. Point base_url to unoblox and drop in your ub-gw key.

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