GPT-5 mini API in India

GPT-5 mini is OpenAI's cost-efficient tier of the GPT-5 family, and on unoblox it is billed the way most Indian engineering and finance teams actually want infrastructure billed: in rupees, on a single monthly GST invoice, through the same OpenAI-compatible endpoint used for every other model in the catalogue. There is no separate OpenAI billing account to open, no international card to attach, and no manual conversion math to reconcile every month.

Why GPT-5 mini instead of full GPT-5

GPT-5 mini sits below GPT-5 in the same model family, aimed at workloads that run at high volume rather than at the hardest reasoning problems — support-ticket drafting, summarisation, RAG answer generation, structured extraction, and chat features where latency and per-call cost matter more than squeezing out the last bit of reasoning depth. Because it shares the GPT-5 lineage, prompts and tool schemas built for GPT-5 typically carry over with only prompt-level tuning, so it's a natural first stop when a GPT-5 workload is proving too expensive to run at scale.

₹ pricing for GPT-5 mini

Model	Input (₹ / 1M tokens)	Output (₹ / 1M tokens)
GPT-5 mini	25.2	201.6
GPT-5	126	1008
GPT-4.1	201.6	806.4

GPT-5 mini's input price is roughly a fifth of GPT-5's, which is why teams route high-frequency, lower-complexity calls to it and reserve full GPT-5 for the harder subset of requests. Treat these as indicative, live rates — the current number always sits on the model page.

Calling GPT-5 mini from the unoblox endpoint

Point any OpenAI SDK — or a raw HTTPS call — at the unoblox base URL and pass the model id. No other code changes are needed if you're migrating an existing OpenAI integration.

curl https://api.unoblox.ai/v1/chat/completions \
  -H "Authorization: Bearer ub-gw-xxxxxxxxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-5-mini",
    "messages": [{"role": "user", "content": "Draft a two-line reply to a shipping delay query."}]
  }'

Streaming, function/tool calling and JSON mode all work exactly as they do against OpenAI directly — only the base URL, the key, and the billing currency change.

What "GPT-5 mini API in India" actually means

GPT-5 mini itself still runs on OpenAI's own infrastructure — unoblox does not move that inference into an Indian data centre, and it would be misleading to claim otherwise. What genuinely changes for an Indian company is the commercial layer: you pay in rupees, you receive one monthly GST invoice from an Indian entity with input tax credit claimable, you don't need an international card, and the same ub-gw- key that calls GPT-5 mini also calls Claude, DeepSeek, Qwen, Llama and every other model behind the same endpoint.

Frequently asked questions

Is GPT-5 mini available in India through unoblox? Yes. Any account can call openai/gpt-5-mini on https://api.unoblox.ai/v1 immediately after signup, billed in rupees against wallet balance.

How is GPT-5 mini priced on unoblox? ₹25.2 per 1 million input tokens and ₹201.6 per 1 million output tokens, charged per token actually used. Rates are indicative and the live figure is always shown on the model page.

Does GPT-5 mini run on servers located in India? No. It runs on OpenAI's infrastructure. unoblox adds ₹ billing, a GST invoice and a single endpoint — not data residency — for this model.

Can I use the same API key for GPT-5 mini and other models? Yes. One ub-gw- key authenticates every model on the platform; you only change the model field between calls.

Is GST input tax credit available on GPT-5 mini usage? Yes. Usage is invoiced monthly by an Indian entity, and that GST is claimable as input tax credit under normal conditions.

Do streaming and tool calling work with GPT-5 mini here? Yes. Both work exactly as they do against OpenAI's own endpoint — only the base URL and key change.

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