# MegaLLM

> MegaLLM is a reliable, unified AI API gateway that gives developers access to 70+ LLMs
> (GPT-5, Claude Opus 4.7, Gemini 3, Grok, Llama, Mistral, and more) through a single
> OpenAI-compatible API — with automatic cross-provider failover and real-time analytics.

## What is MegaLLM?

MegaLLM is an AI API gateway for developers and teams who want one integration instead of many.
Rather than wiring up OpenAI, Anthropic, Google, xAI, Meta, Mistral, and a dozen other providers
separately, you call MegaLLM's single unified API — fully compatible with the OpenAI SDK. Every
request is routed to the best available provider for cost, speed, and reliability, and automatically
fails over if a provider errors or rate-limits.

## Who it's for

Developers and teams already spending on LLM APIs who want lower cost, fewer rate-limit failures,
and provider redundancy — without rewriting their app.

## Key Features

- **One API for 70+ models**: GPT-5, Claude Opus 4.7 / Sonnet / Haiku, Gemini 3, Grok, Llama,
  Mistral, Command R+, DeepSeek, Qwen, and more — through a single endpoint.
- **OpenAI-compatible**: change your base URL and API key; existing OpenAI SDK code works as-is.
  Bidirectional Anthropic API translation (streaming, thinking blocks, tool use) is also supported.
- **Waypoint adaptive routing & failover**: MegaLLM's routing engine scores every provider on
  cost, latency, error rate, and capacity in real time, then routes each request to the best one.
  If a provider fails or rate-limits, it reroutes automatically — your users never see a 503.
- **`model: "auto"`**: an ML classifier analyzes each prompt across 13 task categories and picks
  the most cost-effective model that can handle it.
- **Real-time analytics & cost controls**: track every token, latency (P50/P95/P99), and dollar
  across all models; set spend limits per model, project, or team.
- **Up to 60% cheaper** than calling providers directly. Actual savings vary by workload because
  requests are routed across providers with different underlying costs.
- **High availability** by design, with automatic failover across providers.

## Supported model providers

OpenAI, Anthropic, Google, xAI (Grok), Meta (Llama), Mistral AI, Cohere (Command R+), DeepSeek,
Qwen, and more. New models are typically added within 24–48 hours of release.

See the live, always-current model list at https://megallm.io/models

## Quick start

### Python (OpenAI SDK)

```python
from openai import OpenAI

# Just change base_url and api_key — everything else stays the same
client = OpenAI(
    base_url="https://ai.megallm.io/v1",
    api_key="your-megallm-api-key"
)

response = client.chat.completions.create(
    model="gpt-5",  # or claude-opus-4-7, gemini-3, grok, etc.
    messages=[{"role": "user", "content": "Hello, world!"}]
)

print(response.choices[0].message.content)
```

### JavaScript / TypeScript (OpenAI SDK)

```typescript
import OpenAI from 'openai';

const client = new OpenAI({
  baseURL: 'https://ai.megallm.io/v1',
  apiKey: 'your-megallm-api-key'
});

const response = await client.chat.completions.create({
  model: 'claude-opus-4-7',
  messages: [{ role: 'user', content: 'Hello, world!' }]
});

console.log(response.choices[0].message.content);
```

### cURL

```bash
curl https://ai.megallm.io/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer your-megallm-api-key" \
  -d '{
    "model": "gpt-5",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'
```

## Pricing

MegaLLM offers simple monthly plans plus pay-as-you-go:

- **Basic — $10/month**: all 70+ models, a daily usage allowance, works with Claude Code, Cursor,
  Cline, and OpenCode.
- **Premium — $21/month**: everything in Basic, ~3× the usage, higher-priority access. Most popular.
- **Max — $64/month**: everything in Premium, ~9× the usage, top-priority access — built for teams,
  long agent runs, and heavy automation.
- **Pay-as-you-go**: per-token billing with no minimums, for teams who prefer usage-based pricing.
- **Volume discounts**: available for teams processing 10M+ tokens/month.

Because requests are routed to the most cost-effective available provider, most workloads run up to
60% cheaper than going direct. Pay with Stripe globally or Razorpay in India. No credit card is
required to start.

See full, current pricing at https://megallm.io/pricing

## Why teams choose MegaLLM

- **vs calling providers directly**: automatic failover, one bill, and unified analytics.
- **vs other gateways (OpenRouter, Portkey, LiteLLM)**: managed adaptive routing (Waypoint),
  built-in cost controls, and OpenAI + Anthropic compatibility out of the box.

## Use cases

- **AI application development**: build apps that need several models behind one API.
- **Cost optimization**: route requests to the most cost-effective capable model.
- **Reliability**: automatic failover keeps your app online during provider outages or rate limits.
- **Agents & automation**: works with Claude Code, Cursor, Cline, OpenCode, and any OpenAI-compatible client.
- **Model comparison**: test models side-by-side before committing.

## Links

- Homepage: https://megallm.io
- Pricing: https://megallm.io/pricing
- Models list: https://megallm.io/models
- Documentation: https://docs.megallm.io
- API reference: https://docs.megallm.io/en/api-reference
- News & updates: https://megallm.io/news
- Status page: https://status.megallm.io
- Dashboard: https://megallm.io/dashboard

## Company

- **Legal entity**: Ghostlytics Payments Pvt. Ltd. (d/b/a MegaLLM.io)
- **Registered office**: Bengaluru, Karnataka, India
- **Support email**: support@megallm.io
- **X / Twitter**: https://x.com/megallmio
- **Discord**: https://discord.gg/Megallm
- **GitHub**: https://github.com/Megallm
- **LinkedIn**: https://www.linkedin.com/company/megallm

## Getting started

1. Sign up at https://megallm.io/auth/signup
2. Get your API key from the dashboard
3. Set your base URL to `https://ai.megallm.io/v1`
4. Start calling 70+ models immediately — no credit card required to start.