An OpenRouter alternative

One API for every model.
Topped up in crypto.

GPT, Claude, Gemini and Llama behind a single OpenAI-compatible endpoint. Fund your balance with USDT, USDC or ETH — no credit card, no bank, no monthly minimum.

See pricing

Free to join the waitlist · No wallet connected yet

switch.py
# Already using the OpenAI SDK? Change two lines.
from openai import OpenAI

client = OpenAI(
    base_url="https://api.lmtunnel.xyz/v1",
    api_key="lmt_your_key_here",
)

resp = client.chat.completions.create(
    model="anthropic/claude-sonnet-4",
    messages=[{"role": "user",
               "content": "Hello!"}],
)
Same SDK, same request shape. Swap the model string to change provider.

Routes to

OpenAI GPT-4o Anthropic Claude Google Gemini Meta Llama Mistral DeepSeek Qwen

Three steps to a key

No account forms, no sales call, no card on file.

1 · Send crypto

Transfer USDT, USDC or ETH on Ethereum, Polygon or BNB Chain to the address we generate for you.

2 · Get your key

The balance lands and an lmt_ API key is issued immediately. Nothing to verify, nothing to wait for.

3 · Point your SDK

Change base_url in the OpenAI client you already use. Every model is one string away.

Why teams switch

Built for developers who can't — or would rather not — put a Visa on a US inference bill.

Crypto-native billing

Pay in stablecoins from any wallet. No card, no bank account, no currency conversion surprises.

One endpoint, many models

OpenAI-compatible chat completions and embeddings. Switch providers without touching your code.

Private by default

Prompts and completions are never used for training. Edge processing stays ephemeral.

Global edge network

Requests terminate at the nearest of 60+ edge locations, so routing adds latency you won't notice.

Usage you can audit

Live tokens, latency and spend per model and per key. No end-of-month invoice surprises.

Refundable balance

Unused credit is withdrawable back to your wallet. Top up in any amount, whenever you want.

Per-million-token pricing

Upstream cost plus a flat 3% routing margin. No subscription, no seat fees.

GPT-4o mini
$0.15 /1M in
$0.60 / 1M out
  • 128K context
  • Multimodal input
  • Structured outputs
Llama 4 Scout
$0.08 /1M in
$0.20 / 1M out
  • 10M context MoE
  • Open weights
  • Cheapest long-doc option

Prices shown are indicative launch pricing and may change before general availability.

Ready to build without a card?

Join the waitlist and we'll send your key the day we open.

Where should we send it?

One email when we launch. Nothing else, ever.