Docs

Connect Ervoila to Cursor, Claude Code, OpenCode, or Codex CLI.

Browse models

Configure your snippets

Values are substituted into every example on this page. Nothing is sent anywhere.

API Key

Shown only once at creation — create one under API Keys, then paste it here.

Model ID

Loading the catalogue…

Base URL

https://loveuforever.freesrv.com/v1

OpenAI-compatible. Anthropic clients use the same host without the /v1 suffix.

Quickstart

cURL

curl https://loveuforever.freesrv.com/v1/chat/completions \
  -H "Authorization: Bearer bps_YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "[MODEL_ID]",
    "messages": [{ "role": "user", "content": "Hello" }]
  }'

Python

from openai import OpenAI

client = OpenAI(
    base_url="https://loveuforever.freesrv.com/v1",
    api_key="bps_YOUR_API_KEY",
)

response = client.chat.completions.create(
    model="[MODEL_ID]",
    messages=[{"role": "user", "content": "Hello"}],
)

print(response.choices[0].message.content)

Node

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://loveuforever.freesrv.com/v1",
  apiKey: process.env.ERVOILA_API_KEY,
});

const response = await client.chat.completions.create({
  model: "[MODEL_ID]",
  messages: [{ role: "user", content: "Hello" }],
});

console.log(response.choices[0].message.content);

Endpoints

Method	Path	Auth	Purpose
POST	/v1/chat/completions	API key	OpenAI-compatible chat completions
POST	/v1/messages	API key	Anthropic-compatible messages
GET	/v1/models	Public	Catalogue with retail pricing
GET	/api/balance	Session	Account balance
GET	/api/usage	Session	Aggregated usage and request history
POST	/api/keys/create	Session	Mint an API key (shown once)
DELETE	/api/keys/:id	Session	Revoke an API key

Client setup

Cursor

Settings → Models → OpenAI-compatible provider

Enable the OpenAI-compatible provider, then paste the Base URL https://loveuforever.freesrv.com/v1 and your API key. Add the model id exactly as it appears in the catalogue.

cURL

curl https://loveuforever.freesrv.com/v1/chat/completions \
  -H "Authorization: Bearer bps_YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "[MODEL_ID]",
    "messages": [{ "role": "user", "content": "Hello" }]
  }'

Claude Code

Environment variables

Claude Code has poor compatibility with non-Claude models. Use Ervoila's Claude models with this client.

Bash / macOS / Linux

export ANTHROPIC_BASE_URL="https://loveuforever.freesrv.com"
export ANTHROPIC_API_KEY="bps_YOUR_API_KEY"
export ANTHROPIC_MODEL="[MODEL_ID]"
claude

PowerShell

$env:ANTHROPIC_BASE_URL="https://loveuforever.freesrv.com"
$env:ANTHROPIC_API_KEY="bps_YOUR_API_KEY"
$env:ANTHROPIC_MODEL="[MODEL_ID]"
claude

OpenCode

Create opencode.json in your project root

opencode.json

{
  "provider": {
    "ervoila": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "Ervoila Router",
      "options": {
        "baseURL": "https://loveuforever.freesrv.com/v1",
        "apiKey": "bps_YOUR_API_KEY"
      },
      "models": {
        "[MODEL_ID]": { "name": "[MODEL_ID]" }
      }
    }
  },
  "model": "ervoila/[MODEL_ID]"
}

Codex CLI

Save as $HOME/.codex/config.toml

config.toml

model = "[MODEL_ID]"
model_provider = "ervoila"

[model_providers.ervoila]
name = "Ervoila Router"
base_url = "https://loveuforever.freesrv.com/v1"
env_key = "ERVOILA_API_KEY"
wire_api = "responses"

Authentication & errors

Presenting your key

Both header styles are accepted

Keys are prefixed bps_. Send one as a bearer token, or as x-api-key for Anthropic-compatible clients — that is what Claude Code and similar tools use.

Headers

Authorization: Bearer bps_YOUR_API_KEY
--- or ---
x-api-key: bps_YOUR_API_KEY

Your key is never forwarded upstream; only the provider's own credential is.

Error codes

What each status means

Bad request: 400 missing or malformed field
Unauthorized: 401 key missing, wrong, or revoked
Payment required: 402 balance or key spend limit exhausted
Forbidden: 403 CAPTCHA check failed
Not found: 404 unknown model or route
Payload too large: 413 request body over the limit
Too many requests: 429 rate limited — retry with backoff
Bad gateway: 502 upstream model provider failed
Unavailable: 503 upstream unreachable
Gateway timeout: 504 upstream exceeded the deadline

Top up your balance to start sending requests — credits never expire and top-ups are 1:1.Top up