Xyris Console
Xyris Console
Access state-of-the-art language models via a clean, OpenAI-compatible API. Build faster with competitive pricing, high throughput, and zero infrastructure overhead.
Every new account is credited with $500.00 in free API usage instantly. Start building immediately, with no billing setup or waiting.
Production-ready models you can call today. Drop in your API key and start generating.
Responses start streaming in milliseconds. Our infrastructure is optimized for first-token speed so your users never wait.
At $0.30 input / $0.50 output per 1M tokens, Xyris models cost a fraction of legacy providers, without sacrificing quality.
One-line migration. Change your baseURL and apiKey, and everything else stays the same. Works with any OpenAI SDK.
Your requests are never logged for training. Per-key access control, usage tracking, and instant revocation built in.
Monitor token consumption, latency, and spend in real time from your developer console, with no third-party dashboards needed.
Requests are routed to the nearest inference node automatically. Consistent performance regardless of where your users are located.
Swap a single line of code in your existing applications to start routing completions through Xyris at a fraction of the cost.
import openai
client = openai.OpenAI(
base_url="https://xyris-api-proxy.vercel.app/v1",
api_key="your-xyris-token-here"
)
response = client.chat.completions.create(
model="XYRIS-FLASH",
messages=[{"role": "user", "content": "Hello, Xyris!"}],
stream=True
)
for chunk in response:
print(chunk.choices[0].delta.content or "", end="")We host optimized inference instances directly at the edge, offering the same intelligence at up to 70% cheaper rates than raw legacy APIs.