Technical Specifications

Model Architecture	Aylar-MoE Dense Transformer (8x22B)
Context Length	131,072 tokens (128k FlashAttention)
Token Generation Rate	~145 tokens / sec per stream
Endpoint Protocol	HTTPS / SSE Streaming (OpenAI Compatible)
SLA Guarantee	99.99% Uptime with Tier 1 Fallback

# Request authentication and execution snippet

$ curl -X POST https://api.aylarllm.io/v1/deploy \

-H "Authorization: Bearer $AYLAR_KEY" \

-H "Content-Type: application/json" \

-d '{"plan": "aylarllm-developer-api", "region": "us-east-1"}'

# Python SDK Quickstart

import aylar

client = aylar.Client(api_key="AYLAR_SECRET_KEY")

deployment = client.nodes.provision(

product_id="9460014874843",

variant="Default Title"

)

print(f"Node online at: {deployment.ip_address}")

// Node.js / TypeScript SDK

import { AylarClient } from '@aylar/sdk';

const aylar = new AylarClient({ apiKey: process.env.AYLAR_KEY });

const res = await aylar.instances.create({

sku: 'aylarllm-developer-api'

});

console.log('Instance ID:', res.id);

INSTANT PROVISIONING Automated Delivery

AylarLLM Developer API

$100.00

Build and test AI-powered applications with AylarLLM Developer API.

Designed for developers who need more usage and flexibility than a basic starter plan.

Best For: Freelancers, indie developers, startups, and growing AI projects.