AI Gateway

Pay for AI with
your VibeNest credits.

Any app you host can use AI without a personal OpenAI account. VibeNest runs an OpenAI-compatible gateway — paste your VibeNest key as the API key, point the app's base URL at us, and your credits are billed at provider cost plus a small margin. No card, no separate AI subscription.

Get your AI key See pricing

Set it up in 3 steps

1
Issue a key on /dashboard/ai-keys — you'll see it once, so copy it.
2
In your app's AI settings, choose provider OpenAI-compatible.
3
Fill in the four fields below, pick a model, and save. Every request is billed to your credits.

Base URL

https://ai.vibenest.net/v1

Provider

OpenAI-compatible

API Key

vng_sk_… (from /dashboard/ai-keys)

Model

google/gemini-2.5-flash (or another from the list)

Or call it directly

It speaks the standard OpenAI Chat Completions API — any OpenAI SDK works, just change the base URL and key.

curl

curl https://ai.vibenest.net/v1/chat/completions \
  -H "Authorization: Bearer vng_sk_..." \
  -H "Content-Type: application/json" \
  -d '{"model":"google/gemini-2.5-flash","messages":[{"role":"user","content":"hi"}]}'

Python

from openai import OpenAI
client = OpenAI(base_url="https://ai.vibenest.net/v1", api_key="vng_sk_...")
client.chat.completions.create(
    model="google/gemini-2.5-flash",
    messages=[{"role": "user", "content": "hi"}],
)

JavaScript

import OpenAI from "openai";
const client = new OpenAI({
  baseURL: "https://ai.vibenest.net/v1",
  apiKey: "vng_sk_...",
});

Available models

Model id	Good for	File / PDF input
`openai/gpt-5.1`	OpenAI flagship	yes
`openai/gpt-5-mini`	cheap + capable	yes
`openai/gpt-5-nano`	cheapest OpenAI	yes
`openai/gpt-4.1-mini`	popular workhorse	yes
`anthropic/claude-opus-4.8`	top quality	yes
`anthropic/claude-sonnet-4.6`	balanced flagship	yes
`anthropic/claude-haiku-4.5`	cheap quality	yes
`google/gemini-2.5-pro`	strong, long context	yes
`google/gemini-2.5-flash`	fast, cheap, document import	yes
`google/gemini-2.5-flash-lite`	ultra-cheap	yes
`x-ai/grok-4.3`	latest Grok	yes
`deepseek/deepseek-chat-v3-0324`	cheapest text-only	no
`deepseek/deepseek-r1`	reasoning (text)	no
`meta-llama/llama-4-maverick`	popular open model	no
`qwen/qwen3-max`	popular open model	no

The live list is always at GET https://ai.vibenest.net/v1/models (needs your key).

How billing works

Each call is billed to your VibeNest credits, at the model's provider cost plus a small margin.
You only pay for what you use (tokens), measured from the provider's own usage report.
Out of credits → the call returns 402 insufficient_quota. Top up here.
If the provider call fails, you're not charged.
Your usage is visible per key on /dashboard/ai-keys.

Using it with Reactive Resume

In Reactive Resume go to Settings → Integrations → Add Provider and enter: provider OpenAI-compatible, Base URL https://ai.vibenest.net/v1, Model google/gemini-2.5-flash (file-capable — needed for PDF/Word import), and your vng_sk_… key. Click Test, then Save.

Reactive Resume stores the AI provider per account, so each user of an instance pastes their own key and pays with their own credits.

Is it safe?

Your key only spends your credits — it grants no access to your account, projects or data.
Keys are stored hashed; the raw value is shown once at issue. Revoke or rotate any time on /dashboard/ai-keys.
If a key leaks (e.g. from a deployed container), the blast radius is bounded — revoke it and issue a new one.