Hormouz AIHormouzSign in

OpenAI: gpt-oss-20b: AI model

gpt-oss-20b is an OpenAI model. An open-weight 21B-parameter Mixture-of-Experts model for lower-latency, single-GPU use, with up to 131K-token context. Reachable from inside Iran through Hormouz.

Context window:
131,072 tokens
Input:
$0.02/1M tokens
Output:
$0.09/1M tokens
Type:
text
  • Tools
  • Streaming
  • Long context

Thinking levels: Always thinks · high · medium · low

gpt-oss-20b is an open-weight 21B-parameter Mixture-of-Experts model for lower-latency, single-GPU use. gpt-oss-20b is an open-weight 21B-parameter model released by OpenAI under the Apache 2.0 license. It uses a Mixture-of-Experts architecture with 3.6B active parameters per forward pass, optimized for lower-latency inference and single-GPU deployability, with support for reasoning-level configuration, tool use and structured outputs. It supports a context window of up to 131K tokens, with training data through Jun 2024.

gpt-oss-20b is served through Hormouz behind an OpenAI-compatible API. It supports tool calling and step-by-step reasoning. Point your base URL at Hormouz and drop in a Hormouz key, and existing OpenAI SDK code runs unchanged.

Pricing is usage-based: $0.029 per million input tokens and $0.14 per million output tokens. Hormouz provides stable access to the model from inside Iran, with no foreign credit card or circumvention tooling required.

gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license. It uses a Mixture-of-Experts (MoE) architecture with 3.6B active parameters per forward pass, optimized for...

Quickstart

curl https://api.hormouz.net/v1/chat/completions \
  -H "Authorization: Bearer ng-YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-oss-20b",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Specs

VendorOpenAI
TypeText
Context window131,072 tokens
Pricing$0.02/M in · $0.09/M out
APIChat Completions
CapabilitiesStreaming · Tools

Benchmarks

Reasoning

GPQA Diamond
Graduate-level scientific reasoning
61.1%
HLE
Humanity's Last Exam
5.3%
IFBench
Instruction-following
57.8%
AA-LCR
Long-context reasoning
31%
CritPt
Research-level physics reasoning
0%
AA Omniscience
Broad factual knowledge accuracy
15.1%
Non-Hallucination
Rate of avoiding fabricated answers
13.3%

Coding

Terminal-Bench Hard
Agentic coding & terminal use
4.5%

Agentic

τ²-Bench
Tool-use agents in dual-control scenarios
50.3%

Source: Artificial Analysis

Design Arena

How Design Arena works ↗
Data Viz
Top 93%
940Elo
Win rate: 39.7%Avg time: 8.4s
Website
Top 96%
859Elo
Win rate: 27.9%Avg time: 13.1s

Metrics sourced from Design Arena

Code example

cURL
curl https://api.hormouz.net/v1/chat/completions \
  -H "Authorization: Bearer ng-YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-oss-20b",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'
Python
from openai import OpenAI
client = OpenAI(api_key="ng-YOUR_API_KEY", base_url="https://api.hormouz.net/v1")
response = client.chat.completions.create(
    model="openai/gpt-oss-20b",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)
JavaScript
import OpenAI from "openai";
const client = new OpenAI({ apiKey: "ng-YOUR_API_KEY", baseURL: "https://api.hormouz.net/v1" });
const response = await client.chat.completions.create({
  model: "openai/gpt-oss-20b",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);

Buy the OpenAI: gpt-oss-20b API, paid in rial

  • Buy credit with an Iranian bank card; no foreign card or account needed.
  • One Hormouz key reaches OpenAI: gpt-oss-20b and 600+ other models, and you pay only for what you use.
  • No VPN and no risk of a banned account; just change the base URL to api.hormouz.net/v1.
Create an accountBuy ChatGPT (OpenAI) API access, paid in rial

Use via Hormouz

Connect to this model and 600+ others with one API key, then see usage down to each request.

Request company access

FAQ

How much does gpt-oss-20b cost?
Billing is usage-based: $0.029 per million input tokens and $0.14 per million output tokens. There is no fixed monthly fee. You pay only for what you use, and current rates are shown in the Hormouz dashboard.
How do I access gpt-oss-20b from inside Iran?
Create an account on Hormouz and generate an API key. Hormouz provides a stable path to gpt-oss-20b from inside Iran, with no foreign credit card or circumvention tooling required.
Is gpt-oss-20b compatible with OpenAI code?
Yes. Hormouz exposes an OpenAI-compatible API. Just change the base URL to Hormouz and swap in your Hormouz key. Your existing SDKs and chat-completions code run without a rewrite.
What context window and capabilities does gpt-oss-20b have?
gpt-oss-20b supports a context window of up to 131K tokens. It also supports tool calling and step-by-step reasoning. That makes it a good fit for lower-latency inference, on-device or single-GPU deployment, and agentic tasks such as function calling and tool use.
What is gpt-oss-20b good for?
gpt-oss-20b works well for lower-latency inference, on-device or single-GPU deployment, and agentic tasks such as function calling and tool use.
What is gpt-oss-20b's knowledge cutoff?
gpt-oss-20b's training data runs through Jun 2024. For information newer than that, provide the relevant context in your prompt or connect a tool that fetches live data.

Updated 2026-10-13