The routing layer for production AI

One API.
Every model.
Routed right.

RouterStack sits between your app and 300+ AI models. One OpenAI-compatible endpoint, one key, one bill — with cost-aware routing and automatic failover when a provider goes down.

$ base_url = "https://api.routerstack.ai/v1" — that's the whole migration.
ROUTING TABLE — us / eu / apac● LIVE
your app POST /v1/chat routerstack cost · latency · uptime gpt-5.2openai · 41ms claude-opusanthropic · 38ms llama-4-70brouterstack gpu · 22ms deepseek-v4deepseek · 63ms mistral-largerate-limited → rerouted
healthy route selected · cheapest within SLO down — traffic rerouted, zero 5xx to you
Quickstart

Change one line. Keep your SDK.

RouterStack speaks the OpenAI wire format. Point your existing OpenAI SDK at our base URL and every model in the catalog becomes a string you can swap at runtime.

 app.py
from openai import OpenAI

client = OpenAI(
    base_url="https://api.routerstack.ai/v1",
    api_key="rs-...",  # one key for everything
)

resp = client.chat.completions.create(
    model="anthropic/claude-opus",  # or any of 300+
    messages=[{"role": "user",
               "content": "Hello"}],
    extra_body={
        "route": "cheapest",      # or "fastest"
        "fallbacks": ["openai/gpt-5.2",
                      "meta/llama-4-70b"],
    },
)
01

Swap the base URL

Set base_url to api.routerstack.ai/v1. Streaming, tool calling, JSON mode, vision — all pass through unchanged.

02

Pick models as strings

Model selection becomes a runtime decision, not an architecture decision. A/B a new model by changing one string.

03

Declare your fallbacks

If a provider is down or rate-limited, the request reroutes mid-flight to your next choice. Your users never see the outage.

04

Watch one dashboard

Tokens, spend, latency, and error rates per model, per key, per app — in one place instead of nine provider consoles.

Catalog

300+ models. One namespace.

Frontier models from every major lab, plus open-weight models served on our own GPU fleet — RTX Pro 6000 Blackwell and H200 nodes — often at the lowest per-token price in the catalog.

ModelProviderContextInput / Output per 1M
openai/gpt-5.2OpenAI400K$1.25 / $10.00
anthropic/claude-opusAnthropic200K$5.00 / $25.00
google/gemini-3-proGoogle1M$1.25 / $10.00
meta/llama-4-70bRouterStack GPU128K$0.18 / $0.42
deepseek/deepseek-v4DeepSeek160K$0.26 / $1.10
qwen/qwen3-235bRouterStack GPU128K$0.20 / $0.60
mistral/mistral-large-3Mistral128K$1.80 / $5.40
Illustrative launch catalog. Live pricing is published per-model at launch and billed on each model's native tokenizer.
Why a routing layer

Built for the day a provider goes down.

Every team that ships AI to production eventually learns the same lesson: single-provider is a single point of failure. RouterStack makes multi-provider the default, not a rewrite.

[FAILOVER]

Automatic failover

Provider 529s, rate limits, and timeouts trigger instant rerouting to your declared fallbacks. Zero code in your app, zero downtime for your users.

[COST]

Cost-aware routing

Route "cheapest" sends each request to the lowest-price provider that meets your latency SLO. Teams typically cut inference spend 30–60% without touching quality.

[SPEED]

Latency-first option

Route "fastest" picks the provider with the best live p50 from your region, measured continuously — not from a stale status page.

[POLICY]

Data policies per key

Restrict which prompts may reach which providers. Pin sensitive workloads to zero-retention endpoints or to our own GPU fleet only.

[OBSERVE]

Full observability

Per-request logs, token counts, spend, and latency histograms across every model — exportable, with spend alerts before the bill surprises you.

[STACK]

Our own metal underneath

Open-weight models run on RouterStack's own GPU infrastructure in tier-3 data centers — not resold capacity. When we quote a price, we control the stack behind it.

Pricing

Pay for tokens. Nothing else.

No subscription, no platform fee at launch, no minimums. Buy credits, spend them on any model in the catalog.

Developers

Pay as you go

$0 /month + model token rates
  • All 300+ models, one key
  • Credits never expire
  • Free tier of open-weight models to start
  • Community support
Teams & Enterprise

Committed use

Custom — volume discounts
  • Invoiced billing & SSO
  • Private routing policies & audit logs
  • Dedicated capacity on our GPU fleet
  • 99.9% routed-uptime SLA
Early access

Get your API key first.

We're onboarding in batches. Early access teams get founder pricing locked for 12 months and $50 in launch credits.

✓ You're on the list. We'll email your key when your batch opens.