Hormouz AIHormouzSign in

Google: Gemini 3 Flash Preview: AI model

Gemini 3 Flash Preview is a Google multimodal model for reasoning, coding and everyday chat, with a 1.0M-token context window. Available via Hormouz with stable access from inside Iran.

Context window:
1,048,576 tokens
Input:
$0.5/1M tokens
Output:
$3/1M tokens
Type:
text, image, audio, video
  • Vision
  • Tools
  • Streaming
  • Long context
Good for:Popular for Finance · #4Popular for Health · #8Popular for Legal · #9Ranks are based on global usage data

Thinking levels: high · medium · low · minimal

Gemini 3 Flash Preview is a Google Gemini-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 accepts text, images, PDFs and files, audio and video and returns text, with a context window of up to 1.0M 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 $0.5/M in · $3/M out. Gemini 3 Flash Preview is served through Hormouz behind an OpenAI-compatible API, with stable access from inside Iran.

Gemini 3 Flash Preview is a high speed, high value thinking model designed for agentic workflows, multi turn chat, and coding assistance. It delivers near Pro level reasoning and tool...

Quickstart

curl https://api.hormouz.net/v1/chat/completions \
  -H "Authorization: Bearer ng-YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "google/gemini-3-flash-preview",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Specs

VendorGoogle
TypeText, Image, Audio, Video
Context window1,048,576 tokens
Pricing$0.50/M in · $3.00/M out
APIChat Completions
CapabilitiesStreaming · Vision · Tools

Benchmarks

Reasoning

GPQA Diamond
Graduate-level scientific reasoning
89.8%
HLE
Humanity's Last Exam
36.6%
IFBench
Instruction-following
78%
AA-LCR
Long-context reasoning
78%
CritPt
Research-level physics reasoning
8.6%
AA Omniscience
Broad factual knowledge accuracy
53.4%
Non-Hallucination
Rate of avoiding fabricated answers
7%

Coding

Terminal-Bench Hard
Agentic coding & terminal use
38.6%

Agentic

τ²-Bench
Tool-use agents in dual-control scenarios
80.4%

Source: Artificial Analysis

Design Arena

How Design Arena works ↗

Ranking distribution

First
18,772
Second
17,770
Third
1,340
Fourth
1,181

10880 tournaments ?

Agentic Slides
Top 89%
1,073Elo
Win rate: 39.3%
Agentic Slides (Python)
Top 89%
1,075Elo
Win rate: 39.3%
Android Native
Top 89%
1,065Elo
Win rate: 48.1%
Full-Stack
Top 77%
1,061Elo
Win rate: 47.4%
Godot Game Dev
Top 53%
1,161Elo
Win rate: 50.6%
Mobile Apps
Top 89%
1,087Elo
Win rate: 46.6%
Python Slides
Top 96%
1,036Elo
Win rate: 38.3%Avg time: 241s
Web Apps
Top 81%
1,103Elo
Win rate: 48.5%
3D
Top 37%
1,194Elo
Win rate: 62.7%Avg time: 24.4s
Code
Top 39%
1,195Elo
Win rate: 57.6%Avg time: 34.5s
Game Dev
Top 47%
1,179Elo
Win rate: 58.3%Avg time: 35.6s
Website
Top 39%
1,203Elo
Win rate: 57%Avg time: 35.8s

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": "google/gemini-3-flash-preview",
    "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/gemini-3-flash-preview",
    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/gemini-3-flash-preview",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);

Buy the Google: Gemini 3 Flash Preview API, paid in rial

  • Buy credit with an Iranian bank card; no foreign card or account needed.
  • One Hormouz key reaches Google: Gemini 3 Flash Preview 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 Gemini 3 Flash Preview accept?
It accepts text, images, PDFs and files, audio and video and returns text. Reasoning effort ("thinking") is configurable.
What is the context window of Gemini 3 Flash Preview?
Up to 1.0M tokens, suitable for very long documents and large codebases.
How much does Gemini 3 Flash Preview cost?
Usage-based pricing: $0.5/M in · $3/M out.
How does Gemini 3 Flash Preview 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 Gemini 3 Flash Preview 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