High-Speed AI Gateway + OpenAI & Anthropic Compatible

Locked-Down AI API Gateway with Exact Micro-USD Billing

Access frontier and high-speed AI models through drop-in OpenAI and Anthropic endpoints, atomic credit billing, and full bilingual control.

9 Active AI Models
~220ms Hot Stream TTFT
$0.000001 Atomic Ledger
OpenAI + Claude Native Wire SDKs
Drop-in compatible with:
OpenAI SDK Anthropic SDK Cursor Cline LangChain Vercel AI
LOPI QUANTUM CORE // 3D Drag 3D Reactor to Rotate
GLM 5.3 · FRONTIER DeepSeek V4 Pro SSE Stream · ~220ms Micro-USD Vault
ARCHITECTURE & SECURITY

Why LOPI — Locked API

ZERO-LEAK SECURITY

Zero-Leak Private Inference

Every AI request runs over encrypted isolated channels with zero external data leakage.

Argon2id Keys Strict CSP Isolated Inference
1 USD = 1,000,000 µUSD

Atomic Micro-USD Ledger

Every request uses two-phase hold and settlement arithmetic with an immutable append-only audit trail.

DUAL WIRE PROTOCOL

OpenAI & Anthropic Drop-In

Use /v1/chat/completions and /v1/messages with streaming SSE and standard SDKs.

/v1/chat/completions /v1/messages
ENTERPRISE GOVERNANCE

4-Role Governance & Resellers

Strict server-side RBAC for User, Reseller, Admin, and Superadmin with tenant isolation.

USER RESELLER ADMIN IMMUTABLE AUDIT
DEVELOPER WORKFLOW

How It Works

01

Generate Scoped Key

1. Register, verify your email, and activate a plan or top-up credits.

02

Point Any SDK Endpoint

2. Generate a one-time sk-lopi API key with optional spend and rate limits.

03

Stream & Settle Atomically

3. Call /v1/chat/completions or /v1/messages and track real-time usage and ledger entries.

LIVE INFERENCE CATALOG

Available Models

Live active models currently served by the LOPI gateway

FRONTIER SSE Streaming

glm-5.3

GLM 5.3 high-speed general & coding intelligence model.

FRONTIER SSE Streaming

deepseek-v4-pro

DeepSeek V4 Pro flagship analytical & reasoning model.

TURBO SSE Streaming

deepseek-v4.1-flash

DeepSeek V4.1 Flash ultra-low-latency streaming model.

CODING SSE Streaming

qwen2.5-coder:1.5b

Qwen 2.5 Coder 1.5B instant code generation & completion model.

REASONING SSE Streaming

deepseek-r1:1.5b

DeepSeek R1 1.5B fast chain-of-thought reasoning model.

FRONTIER SSE Streaming

gpt-4o-mini

GPT-4o Mini drop-in compatible fast chat & coding endpoint.

FRONTIER SSE Streaming

claude-3-5-haiku-latest

Claude 3.5 Haiku drop-in compatible fast assistant endpoint.

CODING SSE Streaming

qwen2.5-coder:7b

Qwen 2.5 Coder 7B software engineering & architecture endpoint.

REASONING SSE Streaming

deepseek-r1:8b

DeepSeek R1 8B structured logic & reasoning endpoint.

PREDICTABLE SCALE

Simple Subscription Plans

Predictable USD and IDR pricing configured directly in the platform catalog

Weekly

Weekly Starter

$4 / 7 days

Equivalent: IDR 64,000

  • Credits included: $4.00
  • Up to 30 RPM · 30000 TPM
  • OpenAI & Anthropic SSE Ready
Select Plan
Yearly

Yearly Scale

$100 / 365 days

Equivalent: IDR 1,600,000

  • Credits included: $100.00
  • Up to 120 RPM · 120000 TPM
  • OpenAI & Anthropic SSE Ready
Select Plan
INSTANT INTEGRATION

API Compatibility Examples

OpenAI Chat Completions
curl -X POST https://lopi.my.id/v1/chat/completions \
  -H "Authorization: Bearer sk-lopi-demo" \
  -H "Content-Type: application/json" \
  -d '{"model":"glm-5.3","stream":true,"messages":[{"role":"user","content":"Hello"}]}'
Anthropic Messages API
curl -X POST https://lopi.my.id/v1/messages \
  -H "x-api-key: sk-lopi-demo" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{"model":"deepseek-v4-pro","max_tokens":256,"messages":[{"role":"user","content":"Hello"}]}'
TECHNICAL FAQ

Frequently Asked Questions

Q1

What happens if an upstream AI call fails mid-stream?

Unused reserved credit holds are automatically released back to your available balance without double-charging.

Q2

Can I switch between English and Bahasa Indonesia?

Yes, every page and API error supports full English and Bahasa Indonesia localization.

Q3

Can I set spend limits and allowed models per API key?

Yes. Each sk-lopi key supports optional spend caps, per-key RPM/TPM rate limits, expiration dates, and model allowlists.

Q4

How are local models isolated from public networks?

Local inference runs exclusively on an isolated KVM node protected by strict UFW firewall rules accessible only from the application gateway.

Ready to route Frontier AI through one locked gateway?

Issue scoped sk-lopi keys in seconds with real-time SSE streaming and exact micro-USD accounting.