Now in private beta

One API for
Every Major LLM

Change base URL. Get routing, billing, rate limits, and observability in one place.

Request early access

Features

Everything between your app and the model

Aramm handles routing, billing, and safety defaults so you can focus on your product.

Observability

See what each call costs before the bill arrives.

Track tokens, cost, latency, and errors by model and API key. Spot spikes early and export data when you need it.

Governance

Caps and allowlists that actually enforce.

Per-key rate limits, model allowlists, and spend caps applied on every request, not just in the dashboard.

Guardrails

Catch sensitive data before it leaves your app.

Detect common Singapore patterns (NRIC, phone numbers, payment cards). Block the request, redact the value, or log and allow per API key.

Model Catalog

One base URL. Many models.

Call any model through using Aramm Gateway. Swap model in your request, no new SDK.

One Gateway. Many Model Families.

The models you already use, routed through a Singapore-hosted gateway. Same request format. We handle the provider wiring.

Anthropic
Claude — strong for coding, reasoning, and long context
OpenAI
GPT models — general purpose and multimodal

Integration

Change two lines.
Keep everything else.

Works with the OpenAI Python/JS SDK, cURL, and anything else that speaks the Chat Completions API.

from openai import OpenAI

client = OpenAI(
    api_key="sk-arm-v1-YOUR_KEY",
    base_url="https://gateway.aramm.ai/v1",
)

response = client.chat.completions.create(
    model="claude-sonnet",
    messages=[
        {"role": "user", "content": "Explain neural networks simply."}
    ],
    stream=True,
)

for chunk in response:
    print(chunk.choices[0].delta.content or "", end="")

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "sk-arm-v1-YOUR_KEY",
  baseURL: "https://gateway.aramm.ai/v1",
});

const stream = await client.chat.completions.create({
  model: "claude-sonnet",
  messages: [
    { role: "user", content: "Explain neural networks simply." }
  ],
  stream: true,
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content ?? "");
}

curl https://gateway.aramm.ai/v1/chat/completions \
  -H "Authorization: Bearer sk-arm-v1-YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet",
    "messages": [
      {
        "role": "user",
        "content": "Explain neural networks simply."
      }
    ],
    "stream": true
  }'

Compatibility

Works with the tools you already use.

Use Aramm Gateway with coding agents, workflow tools, and frameworks. Change the endpoint; keep your workflow.

Claude Code
Codex
OpenCode
Hermes
Pi
Cursor
Cline
LangGraph

Simple to adopt. Transparent by design.

We won't claim compliance you still have to implement yourself. We won't hide beta limitations. If something isn't ready, we'll say so.

Get started

Live in minutes, not months.

No servers to run. No separate Anthropic and OpenAI billing. API key + base URL.

Create an account

Top up credits

Add credits via Stripe. Balance is tracked in real time. Set low-balance alerts or enable auto top-up.

Generate an API key

Create a key scoped to your workspace. Set per-key rate limits, model allowlists, and spend caps.

Swap your base_url

Point your base URL to gateway.aramm.ai/v1. Every model, one key, zero code changes.

Questions, answered plainly.

What is Aramm and who is it for?

A managed API in front of multiple LLM providers. Built for indie devs, startups, and engineers who want one key and one bill instead of juggling provider dashboards.

Is Aramm PDPA compliant?

We’re built with Singapore data-protection expectations in mind: Singapore hosting, optional content logging off by default, and PII guardrails. Your app’s compliance is still your responsibility; we document what the platform enforces.

Do you store my prompts or responses?

Content logging is off by default. You can opt in per API key, and when you do, PII redaction runs before anything is stored. You control the retention period: 30 days on Free, configurable up to custom retention on Enterprise.

Which providers are supported?

Anthropic, OpenAI, and open-weight models today. More providers on the way.

How does governance work across teams and API keys?

Set model access, spend limits, and roles once per team or key. Aramm enforces them on every request, with a full audit log of who did what and when. No one can route around a policy from the client side.

Can I use Aramm with LangChain, LangGraph, or other agent frameworks?

Yes. Any framework that accepts a base URL and an OpenAI compatible API key works: LangChain, LangGraph, LlamaIndex, n8n, Dify, and custom agent loops.

How does pricing work? Are there seat fees?

No seat fees, no subscriptions. Top up credits in USD and pay only for what you use. Aramm adds a 5% platform fee when you top up credits; inference runs at provider cost. Credits never expire.

How do I get started?

Request early access, top up credits, generate an API key, and change your SDK's base_url. No provider contracts to sign, no infrastructure to provision. Most teams are live in under five minutes.

Early access

Built for developers in Singapore and ASEAN

We're in private beta. Join the waitlist for early access.

No spam. No marketing emails. Just a heads-up when you're in.

One API forEvery Major LLM