Quickstart
From signup to your first routed request in under 2 minutes.
HiWay2LLM is a smart routing proxy that sits between your code and any LLM provider. You change one line in your existing app — your base_url — and HiWay routes each request to the cheapest model that can still answer it. Average savings: 40 to 85% on real workloads.
1. Create an account
Head to /auth, sign up with an email and a password, and copy the API key shown once. The key starts with hwy_. Save it somewhere safe — we never store it in clear, only its SHA-256 hash, so we cannot recover it if you lose it.
2. Buy your first credits
Go to Dashboard → Billing. Type any amount between $10 and $1,000 and click Proceed to checkout. The first $10 of cumulative credits per user are fee-free — a small welcome from us. Above that, a flat 5% platform fee is added at purchase time.
No markup on tokens
Once you have credits, every request is debited at the exact price the LLM provider charges. There is no per-token markup on top of the 5% upfront fee.
3. Send your first request
HiWay speaks the standard chat-completions wire protocol. Any client that targets it works — OpenAI Python, Anthropic via base URL override, LangChain, Vercel AI SDK, n8n, curl, you name it. Just point your client at our base URL and use your hwy_ key.
from openai import OpenAI
client = OpenAI(
base_url="https://www.hiway2llm.com/v1",
api_key="hwy_YOUR_KEY",
)
response = client.chat.completions.create(
model="claude-haiku",
messages=[{"role": "user", "content": "Say hello in 5 words"}],
)
print(response.choices[0].message.content)
print("Routed to:", response.model)4. See what was routed
Every response carries headers that tell you exactly which model HiWay picked, why, and how fast.
X-Hiway-Model: anthropic/claude-haiku-4-5-20251001
X-Hiway-Tier: light
X-Hiway-Score: 0.05
X-Hiway-Routing-Ms: 0.84Open the Usage tab in the dashboard to see the live timeline of your requests, the cost breakdown by model, and the savings vs. always routing to the most expensive tier.