Chat completions · Docs

The OpenAI-compatible core endpoint: request body, streaming, and response headers.

Endpoint

Drop-in OpenAI-compatible. Base URL: https://router-api.intelli-verse-x.ai/v1. Authenticate with Authorization: Bearer — see Authentication.

POST/v1/chat/completions

Request body

Parameters
Name	Type	Description
`messages`Required	array	OpenAI-format message list: `{ role, content }` objects with `system`, `user`, and `assistant` roles. Vision content parts are supported on tiers with the vision capability.
`model`Optional	string	Tier alias — omit it and we route at your plan's tier. Valid values: `intelliverse`, `intelliverse/air`, `intelliverse/pro`, `intelliverse/pro-max`, `intelliverse/fusion`, a saved preset (`@preset/slug`), or a legacy OpenRouter-style slug (mapped into tier routing).
`stream`Optional	boolean	Stream the response as server-sent events. Defaults to `false`.
`plugins`Optional	array	Per-request plugins, e.g. `[{ "id": "web-search", "max_results": 3 }]`. See Presets & plugins.
`temperature, max_tokens, …`Optional	various	Standard OpenAI sampling and output parameters pass through unchanged.

Example request

curl https://router-api.intelli-verse-x.ai/v1/chat/completions \
  -H "Authorization: Bearer $INTELLIVERSE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "intelliverse",
    "messages": [
      { "role": "system", "content": "You are a concise assistant." },
      { "role": "user", "content": "Summarize this changelog..." }
    ]
  }'

Streaming

Set "stream": true for standard OpenAI server-sent events. Chunk model fields report the tier alias, and the stream terminates with data: [DONE].

JavaScript

const stream = await client.chat.completions.create({
  model: "intelliverse",
  messages: [{ role: "user", content: "Write a haiku about routing." }],
  stream: true,
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content ?? "");
}

Response headers

Every chat response includes credit metering headers:

x-iv-credits-spent — credits consumed by this request.
x-iv-credits-remaining — the app wallet balance after settlement.
x-intelliverse-routing — opt-in routing transparency (send X-Intelliverse-Transparency: routing); see Routing transparency.

Note

Response model fields, SSE chunks, headers, and error messages always report your tier alias — never a provider or model name.

Status codes

Status	Reason	What to do
200	—	Success — standard OpenAI chat-completion body.
402	`credit_exhausted`	Balance is zero. Top up or enable metered overage — see Errors.
403	`tier_required`	The capability needs a higher plan (e.g. Fusion on Pro+).
429	`key_rate_limit`	Honor `Retry-After` — see Rate limits.