Do More AI
Spend Less

Unified access to the world's leading AI models
-faster,cheaper,and built for scale.


import Siray from "siray";
import fs from "node:fs";
const siray = new Siray({ auth: process.env.SIRAY_API_TOKEN });
const model = ""
const input = {
  prompt: "",
  image: siray.loadFromLocal("path/to/image.jpg")
};
const result = await siray.run(model, { input });
const output = result.outputs()[0].url();
console.log(output.url());
fs.writeFile("my-image.jpg", output);


from siray.client import Client
import os
client = Client({ "auth": os.getenv("SIRAY_API_TOKEN") })
model = ""
result = client.run(model,input={
  "prompt": "",
  "image": client.load_from_local("path/to/image.jpg")
})
output = result.outputs()[0]
print(output.url())
with open("my-image.jpg", "wb") as file:
    file.write(output.read())


BASE64_DATA=$(base64 image.jpg | tr -d '\n')
DATA_URL="data:image/jpeg;base64,$BASE64_DATA"
MODEL_URL="https://api.siray.ai/v1/images/generations/async"
curl -X POST $MODEL_URL -d "{
  "prompt": "",
  "image": "$DATA_URL",
  "model": ""
}" -H "Content-Type: application/json" -H "Authorization: Bearer $SIRAY_API_TOKEN"
curl -X GET "$MODEL_URL/$task_id"  
-H "Authorization: Bearer $SIRAY_API_TOKEN"

Do more AI
Spend Less

Unified access to the world's
leading AI models
faster, cheaper, and built for scale.

Unified access to leading AI models to meet 100% of needs

Unified access to leading AI models to meet 100% of needs

300 +

LLMs supported

unified access through one API

30%

Cost savings

reduced by intelligent routing

99.9%

Uptime guarantee

enterprise-level reliability

10ms

overhead latency

fast routing with almost no delay

One API to Access
300
+ Models

Kling 3 Motion Control

Kuaishou

Gemini 3.1 Flash Image Preview (Nano Banana 2)

Google

Gemini 3.1 Pro Preview

Google

Qwen 3.5 Plus

Alibaba

GPT 5.3 CodeX

OpenAI

GLM 5

Z.ai

MiniMax M2.5

MiniMax

Claude Sonnet 4.6 (Prompts ≤ 200K tokens)

Anthropic

Claude Sonnet 4.6 (Prompts > 200K tokens)

Anthropic

Start in 4 Simple Steps.

STEP 1

Create your account and join the next-gen AI Infra platform.

STEP 2

Grab Your API Key

Secure your access to 300+ AI models with one click.

STEP 3

Update Your URL

Integrate Siray.ai into your workflow - no setup friction.

STEP 4

Launch First Call

Run your first model call and experience lightning-fast results.

Build, test, and deploy AI in minutes - not hours.
Connect effortlessly to 300+ high-performance models with lightning-fast,
cost-efficient APIs.

Why choose Siray?

Fast. Reliable. Affordable. Unified.

All Models, One API

Access hundreds of leading AI models through one API - text, image, video, and beyond.

Cost Effective

Siray can save you up to 70% and an average of 30% on API costs while ensuring stability.

Made for Builders

Powerful SDKs, clear docs, and unified endpoints make building with AI effortless.

Instant Response

Experience near-zero delay with intelligent routing and local acceleration for every model request.

Power Everywhere

15+ global data centers deliver reliable, high-performance GPU power - wherever you build.

Built to Last

99.99% uptime, ZDR redundancy, and SOC 2 compliance ensure secure, always-on performance.

Build faster. Scale smarter.

Siray brings together the world's best generative models and GPUs so you can create, test, and deploy without limits. The next era of AI starts with you.

One API to Access300+ Models

Kling 3 Motion Control

Gemini 3.1 Flash Image Preview (Nano Banana 2)

Gemini 3.1 Pro Preview

Qwen 3.5 Plus

GPT 5.3 CodeX

GLM 5

MiniMax M2.5

Claude Sonnet 4.6 (Prompts ≤ 200K tokens)

Claude Sonnet 4.6 (Prompts > 200K tokens)

Start in 4 Simple Steps.

Why choose Siray?

All Models, One API

Cost Effective

Made for Builders

Instant Response

Power Everywhere

Built to Last

One API to Access
300
+ Models