GPT-OSS 20B API in India

gpt-oss 20B is OpenAI's small open-weight reasoning model, available on unoblox as openai/gpt-oss-20b. It is the lowest-priced input option among these new additions, and it bills in rupees with a GST invoice.

What gpt-oss 20B is

OpenAI released gpt-oss 20B in August 2025 under the Apache 2.0 licence. It is a mixture-of-experts model with about 20.9 billion total and 3.6 billion active parameters, a 128k-token context and configurable reasoning effort of low, medium or high. unoblox lists a context length of 131,072 tokens, with reasoning and tool calling, text in and text out.

It is an older and smaller release than the GPT-6 line, which makes it a budget reasoning option rather than a frontier one.

Where it fits

• Cheap reasoning steps for routing, validation and short analyses.
• Tool-calling prototypes where cost per call must be tiny.
• Batch processing of large text volumes.

Use it for cost-sensitive volume. If answers are not good enough, move up to GPT-6 Luna for the same vendor's newer small model, or a larger open model.

₹ pricing for gpt-oss 20B

On unoblox, gpt-oss 20B costs ₹2.89 per million input tokens and ₹13.48 per million output tokens. That displayed price is what you are billed per token; nothing is added on top. The model's context length on unoblox is 131,072 tokens.

Model	Input (₹ / 1M tokens)	Output (₹ / 1M tokens)
gpt-oss 20B	₹2.89	₹13.48
Qwen3.6 35B A3B	₹9.63	₹91.47
Nemotron 3.5 Lightning	₹5.78	₹15.41
GPT-6 Luna	₹9.63	₹48.14

As a worked example, a month with 20 million input tokens and 4 million output tokens comes to about ₹111.72 on gpt-oss 20B (₹57.80 for input plus ₹53.92 for output). Live rates are always on the model page, and each request is rounded up to the nearest paisa.

Calling gpt-oss 20B from the unoblox endpoint

unoblox is OpenAI-compatible, so any OpenAI SDK works by changing the base URL and key.

curl https://api.unoblox.ai/v1/chat/completions \
  -H "Authorization: Bearer YOUR_UNOBLOX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-oss-20b",
    "messages": [{"role": "user", "content": "Summarise this support ticket in two lines."}],
    "max_tokens": 512
  }'

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_UNOBLOX_API_KEY",
    base_url="https://api.unoblox.ai/v1",
)
resp = client.chat.completions.create(
    model="openai/gpt-oss-20b",
    messages=[{"role": "user", "content": "Summarise this support ticket in two lines."}],
    max_tokens=512,
)
print(resp.choices[0].message.content)

Get a key at https://unoblox.ai/sign-in and top up your wallet in rupees.

Notes on parameters, billing and data

Use max_tokens for output limits. This is an open-weight OpenAI model, not an OpenAI-hosted GPT-6 model, so it is not subject to the GPT-6 parameter rules. It runs at an external provider, so we make no India data-residency claim.

Frequently asked questions

Is gpt-oss 20B available in India? Yes, as openai/gpt-oss-20b on https://api.unoblox.ai/v1.

Is it open weight? Yes, under Apache 2.0.

What context length is listed? 131,072 tokens.

Does it support reasoning? Yes, the catalogue lists reasoning and tool calling.

Does it take images? No, it is text in, text out.

Do I get a GST invoice? Yes.

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