Free
$0forever
Explore the gateway with a starter model and instant quota.
- Token cap
- 5M / day
- Rate limit
- 10 req / min
Models includedMistral LargeMistral Medium 3.5Tencent Hy3
Start freeUnifiedAI GatewayOne AI gateway
One AI gateway
for every model
Route requests through one OpenAI-compatible endpoint. Track usage, control credits, manage limits, and keep your AI costs predictable.
- Free model included
- No credit card required
- Set up in under 2 minutes
from openai import OpenAI
client = OpenAI(
base_url="https://api.rerouter.dev/v1",
api_key="sk-rr-xxxxxxxx"
)
response = client.chat.completions.create(
model="auto/rerouter",
messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)Route requests across
top AI providers.
Switch providers without changing your integration.
How it works
3 steps, start coding
From signup to AI workloads in just a few minutes.
STEP 01
Sign up & generate an API key
Create an account, generate an API key, and start using the service in seconds.
- Create an account in seconds
- Generate an API key from the dashboard
- Free tier quota is active instantly
STEP 02
Set up your favorite tool
Paste the base URL and API key into the tool you already use.
base_url: https://api.rerouter.dev/v1
api_key: sk-rr-xxxxSTEP 03
Start building
Use your favorite models through one endpoint.
GPTClaudeGeminiDeepSeekLlama+ More
Trusted by builders,
by the numbers
Real usage flowing through Rerouter.
0Tokens processed daily
0Requests served daily
0Models online
0Uptime SLA
Compatibility
Works with the tools
you already use
Use Rerouter in your favorite coding agents, AI IDEs, and OpenAI-compatible clients.
[CC][CC]
>_OpenAI SDK
◈VS Code
⌘Cursor
λLangChain
$cURL
{ }Any OpenAI client
∞Your stack
AI models pricing
Choose the plan that
fits you best
Access the best AI models through one endpoint with transparent pricing and reliable performance.
MONTHLY SAVES 17%
Hermes
$19/ month
For side projects and daily coding with Hermes models.
- Token cap
- 10M / day
- Rate limit
- 15 req / min
Models includedHermes V2.5Hermes V2.5 Pro
Buy planHermes Pro
$34/ month
Higher caps for freelancers shipping client work daily.
- Token cap
- 15M / day
- Rate limit
- 50 req / min
Models includedHermes V2.5Hermes V2.5 Pro
Buy planMimo
$99/ month
Mimo family models with ultraspeed variants for agents.
- Token cap
- 40M / day
- Rate limit
- 30 req / min
Models includedMimo V2.5Mimo V2.5 ProMimo Ultraspeed
Buy planDeepSeek
$199/ month
DeepSeek V4 family for heavy reasoning workloads.
- Token cap
- 40M / day
- Rate limit
- 25 req / min
Models includedDeepSeek V4 FlashDeepSeek V4 Pro
Buy planMax
$990/ month
Every frontier model, one plan, team-grade limits.
- Token cap
- 100M / day
- Rate limit
- 120 req / min
Models includedAll 245 modelsPriority routing
Buy planNo subscription
Pay as you go
$0.40per 1M tokens
+ list price
+ list price
For variable workloads that need every model without a monthly commitment.
Add credits- All 245 models at provider list price
- $0.40 platform fee per 1M tokens
- Automatic failover and cost routing
- Credits never expire, top up from $10
- Rate limit 60 req/min per model
How usage pricing works
Prepaid credits are drawn down per request. You pay the provider's list price for tokens, plus a flat $0.40 platform fee per million tokens routed.
1M input tokensprovider list price
1M output tokensprovider list price
Platform fee$0.40 / 1M tokens
Commitmentnone
Need predictable monthly cost with a token cap instead? Switch to the tab.
FAQs
Frequently asked
questions
Quick replies to the questions we get asked the most.
Is Rerouter really OpenAI-compatible?
Yes. Point any OpenAI SDK at our base URL and everything works: chat, streaming, tools, vision and embeddings. Models are namespaced as provider/model.
How does billing work?
Subscriptions give you a daily token cap and rate limit for a fixed daily, weekly or monthly price. Pay as you go charges provider list price plus a $0.40 platform fee per 1M tokens from prepaid credits.
What happens when I hit my token cap?
Requests keep flowing on Pay as you go credits if you have a balance, otherwise they pause until the cap resets the next day. You can raise limits anytime by upgrading your plan.
What happens when a provider goes down?
Continuous health probes detect degradation and failed requests retry against equivalent models, so your users never see a 503.
Do you train on my data?
Never. Prompts and completions are never used for training, and zero-retention mode disables logging entirely.
Rerouter
Start routing in
two minutes.
One key, one base URL, every model. Free tier included, no credit card required.
Start Free