OpenAI & Anthropic-compatible gateway
Your agent doesn’t need a frontier model for every call.
Change one base_url and every request routes to the cheapest model that passes the bar for that task. When only a frontier model will do, we escalate — and the receipt says so.
$1 free credit · no card · 2-min setup
Integration
One line. That’s the whole migration.
Keep your SDK, your prompts, your tool definitions. Swap the base URL and set model to auto.
from openai import OpenAI
client = OpenAI(
base_url="https://api.vynaris.com/v1", # was https://api.openai.com/v1
api_key=os.environ["VYNARIS_API_KEY"],
)Benchmark
The numbers, not the adjectives.
Method: 2,000 real agent tool calls replayed through the router and through the requested model directly; identical prompts, identical pass/fail graders, list prices. Reproduce it yourself →
Sample data: numbers from internal replay; public benchmark repo coming.
How it works
Route down by default. Escalate when it matters.
Point your agent at one endpoint
Swap base_url in your OpenAI or Anthropic client. Nothing else in your stack changes.
We classify, route, and escalate
Each request is scored against benchmark data and sent to the cheapest model that clears the quality bar. Hard requests go straight to a frontier model.
Every response carries a receipt
The served model and exact cost ride back on the response headers:
x-vynaris-served-model: qwen3-8b-ft x-vynaris-cost-usd: 0.0004 x-vynaris-baseline-usd: 0.0018 x-vynaris-saved: 78%
Receipts, not claims
If you can’t verify it, we shouldn’t say it.
Per-response attribution
Every response names the model that served it and what it cost, in the headers and in your ledger. No blended averages, no hiding the routing.
Public stats page coming
Aggregate routing mix, pass rates, and savings across all traffic, published continuously. The same numbers we see are the numbers you see.
Escalation is a feature
When small models aren’t good enough, we route up to the model you asked for and the receipt reads −0%. You never pay in quality for our savings number.
Compute at cost + 1%. That’s the whole model.
See pricing →Who’s behind this
Built by one person. Here’s exactly where this stands.
No logos, no testimonials, no “trusted by” wall — none of that exists yet and I won’t fake it. What exists: a working gateway, a reproducible benchmark you can run against your own traffic, a public ledger of every routing decision, and my email. If the numbers don’t hold on your workload, you’ll see it in the receipts before I do.
— DHRUMIL · [email protected]
Running >$10k/monthin agent traffic? I’ll run your traffic through the benchmark and show you your number.
Book 30 minutes →