Novion API docs

Docs

Quickstart

Three steps: get a key, send a request, read the reply.

Open the Novion Console, sign in, and create an API key under API keys. Copy it — it's shown once.
Save it as an environment variable: export NOVION_API_KEY=nv-sk-…
Send your first request:

curl https://api.novion.currentflo.com/v1/chat/completions \
  -H "Authorization: Bearer $NOVION_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"novion-core","messages":[{"role":"user","content":"Hello, Novion"}]}'

Base URL: https://api.novion.currentflo.com/v1

Authentication

Send your key on every request, either way:

Authorization: Bearer nv-sk-…
# or
x-api-key: nv-sk-…

Keep keys secret. Novion stores only a hash, so a lost key can't be shown again — revoke it in the Console and make a new one.

Models

ID	Name	Good for
`novion-core`	Novion Core	Fast everyday chat, tools and simple tasks (default)
`gypsy-x`	Gypsy X	Reasoning, research, comparisons
`novamind`	NovaMind	People, psychology, communication
`nql-lite`	NQL Lite	Code
`moonshot`	Moonshot	Scenarios and forecasts

Details on the Models page.

Chat completions

POST /v1/chat/completions — OpenAI format. system messages are supported inside messages.

{
  "model": "gypsy-x",
  "messages": [
    {"role": "system", "content": "Answer in 3 bullets."},
    {"role": "user", "content": "Compare solar leasing vs buying"}
  ],
  "max_tokens": 600,
  "temperature": 0.6
}

Response:

{
  "id": "chatcmpl-…", "object": "chat.completion", "model": "gypsy-x",
  "choices": [{"index": 0, "message": {"role": "assistant", "content": "…"}, "finish_reason": "stop"}],
  "usage": {"prompt_tokens": 42, "completion_tokens": 180, "total_tokens": 222},
  "novion": {"engine": "…", "latency_ms": 1840}
}

Messages

POST /v1/messages — Anthropic format, with a top-level system.

{
  "model": "nql-lite",
  "max_tokens": 1024,
  "system": "You write Cloudflare Workers.",
  "messages": [{"role": "user", "content": "Write a hello-world Worker"}]
}

Returns {"type": "message", "content": [{"type": "text", "text": "…"}], "usage": {…}}.

List models

curl https://api.novion.currentflo.com/v1/models -H "Authorization: Bearer $NOVION_API_KEY"

Errors & limits

Status	Meaning	What to do
400	Bad request (unknown model, empty messages, too long, `stream: true`)	Fix the request body.
401	Missing, wrong or revoked key	Check the key in the Console.
429	Daily limit reached for this key	Wait for midnight UTC or use another key.
502	The model is busy or over its limit	Retry in a moment.

Each key gets 300 requests a day, reset at midnight UTC.
max_tokens is capped at 4,096; conversations at 60,000 characters.
Streaming isn't available yet.
If a model is busy, Novion may answer with a smaller engine — novion.engine tells you which.

SDKs

Python · OpenAI SDK

from openai import OpenAI
client = OpenAI(base_url="https://api.novion.currentflo.com/v1", api_key=NOVION_API_KEY)
r = client.chat.completions.create(model="novion-core", messages=[{"role": "user", "content": "Hi"}])
print(r.choices[0].message.content)

Python · Anthropic SDK

import anthropic
client = anthropic.Anthropic(base_url="https://api.novion.currentflo.com", api_key=NOVION_API_KEY)
r = client.messages.create(model="moonshot", max_tokens=800, messages=[{"role": "user", "content": "3 scenarios for my week"}])
print(r.content[0].text)

JavaScript · fetch

const r = await fetch("https://api.novion.currentflo.com/v1/chat/completions", {
  method: "POST",
  headers: { "Authorization": "Bearer " + process.env.NOVION_API_KEY, "Content-Type": "application/json" },
  body: JSON.stringify({ model: "novion-core", messages: [{ role: "user", content: "Hi" }] })
});
console.log((await r.json()).choices[0].message.content);