Hormouz AIHormouzSign in

Google: Gemma 4 26B A4B : AI model

Gemma 4 26B A4B is a Google multimodal model for reasoning, coding and everyday chat, with a 262K-token context window. Available via Hormouz with stable access from inside Iran.

Context window:
262,144 tokens
Input:
$0.08/1M tokens
Output:
$0.26/1M tokens
Type:
text, image, video
  • Vision
  • Tools
  • Streaming
  • Long context
Good for:Popular for Marketing & SEO · #6Popular for Translation · #14Ranks are based on global usage data

Thinking levels:

Gemma 4 26B A4B is a Google Gemma-family model built for reasoning, coding, mathematics and general chat. It supports configurable "thinking" (reasoning effort levels) so you can trade cost and latency for accuracy on hard tasks. It uses a Mixture-of-Experts (MoE) design, activating only a fraction of its parameters per token for efficient inference. It accepts images, text and video and returns text, with a context window of up to 262K tokens.

Its multimodal input makes it suitable for document understanding, chart and image analysis, and workflows that mix text with images and video. With thinking enabled it is stronger on multi-step reasoning; with thinking minimal it responds quickly and cheaply.

Pricing is usage-based at $0.15/M in · $0.6/M out. Gemma 4 26B A4B is served through Hormouz behind an OpenAI-compatible API, with stable access from inside Iran.

Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind. Despite 25.2B total parameters, only 3.8B activate per token during inference — delivering near-31B quality at...

Quickstart

curl https://api.hormouz.net/v1/chat/completions \
  -H "Authorization: Bearer ng-YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "google/gemma-4-26b-a4b-it",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Specs

VendorGoogle
TypeText, Image, Video
Context window262,144 tokens
Pricing$0.08/M in · $0.26/M out
APIChat Completions
CapabilitiesStreaming · Vision · Tools

Benchmarks

Reasoning

GPQA Diamond
Graduate-level scientific reasoning
71.4%
HLE
Humanity's Last Exam
11.5%
IFBench
Instruction-following
45.4%
AA-LCR
Long-context reasoning
42.3%
CritPt
Research-level physics reasoning
0%
AA Omniscience
Broad factual knowledge accuracy
15.5%
Non-Hallucination
Rate of avoiding fabricated answers
7.9%

Coding

Terminal-Bench Hard
Agentic coding & terminal use
25%

Agentic

τ²-Bench
Tool-use agents in dual-control scenarios
40.4%

Source: Artificial Analysis

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": "google/gemma-4-26b-a4b-it",
    "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="google/gemma-4-26b-a4b-it",
    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: "google/gemma-4-26b-a4b-it",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);

Buy the Google: Gemma 4 26B A4B API, paid in rial

  • Buy credit with an Iranian bank card; no foreign card or account needed.
  • One Hormouz key reaches Google: Gemma 4 26B A4B 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 Gemini 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

What inputs does Gemma 4 26B A4B accept?
It accepts images, text and video and returns text. Reasoning effort ("thinking") is configurable.
What is the context window of Gemma 4 26B A4B?
Up to 262K tokens, suitable for long documents and multi-turn conversations.
How much does Gemma 4 26B A4B cost?
Usage-based pricing: $0.15/M in · $0.6/M out.
How does Gemma 4 26B A4B perform on benchmarks?
Independent scores from Artificial Analysis are shown in the Benchmarks section above, grouped by reasoning, coding and agentic tasks.
How do I access Gemma 4 26B A4B from inside Iran?
Create an account on Hormouz and generate an API key. Hormouz provides stable access from inside Iran via an OpenAI-compatible API.

Updated 2026-10-13