VibeLLM | Cheap API for VibeCoding | GPT | Claude | Grok | Gemini | Chinese

[noware]

VibeLLM

Cheap OpenAI-compatible API for GPT, Claude, Grok, and leading Chinese LLMs

One endpoint for Cursor, Codex, VS Code, OpenCode, SDKs, bots, scripts, and AI agents.

--------------

In Brief

Great vibe coding isn't just about using powerful AI models—it's also about having easy, hassle-free access to them. No international credit cards, no subscriptions, no complicated billing, and no endless configuration.

VibeLLM gives you access to GPT, Claude, Grok, and leading Chinese AI models through a familiar OpenAI-compatible API.

Simply set your `base_url`, add your API key, and start using your favorite tools.

* OpenAI-compatible API endpoint
* GPT, Claude, Grok, and Chinese models through a single API
* Compatible with Cursor, Codex, VS Code, OpenCode, and OpenAI/Anthropic SDKs
* Pay only for the tokens you use
* No subscriptions or recurring fees



Website: https://vibellm.store/

Telegram channel: https://t.me/vibellmstore

Technical Support:  https://t.me/vibellmsupp

---
  Promotion: Get 5 Million Tokens for a Review

Spoiler:

Sign up on VibeLLM.store, then send your registered email address to us on Telegram (@vibellmsupp).

We'll credit your account with 5 million tokens so you can try the service for free.

After you've had a chance to use it, simply leave an honest review about your experience.

How it works:

Register an account.
Send your registered email to us on Telegram.
Receive 5 million free tokens.
Test the service with any supported AI model.
Share your honest review.

No subscription required. No hidden conditions—just try the service and tell us what you think.

Available Models

Spoiler:

GPT Models

* GPT-5.5
* GPT-5.4
* GPT-5.4 Mini

Claude Models

* Fable 5
* Sonnet 5
* Sonnet 4.6
* Sonnet 4.5
* Opus 4.8
* Opus 4.7
* Opus 4.6
* Opus 4.5
* Haiku 4.5

Chinese Models

* Kimi K3
* GLM-5.2
* Qwen 3.7 Max / Plus
* Kimi 2.7 Code
* MiniMax M3
* DeepSeek V4 Flash / Pro
* MiMo v2.5 Flash / Pro
* and more

Other Models

* Grok 4.5
* Composer 2.5 Fast
* Nemotron 3 Ultra
* North Mini Code

---

Pricing & Billing

Spoiler:

Starting from $0.13 per 1M tokens.

The price per 1M tokens depends on your top-up amount — the larger the deposit, the lower the effective token price.

Example:

With a $10 top-up, the effective rate is approximately $0.20 per 1M tokens, giving you about 50.32M tokens.

You only pay for the input and output tokens you actually use.

Payment Methods

Spoiler:

* Cryptocurrency
* Apple Pay *(coming soon)*
* Google Pay *(coming soon)*
* Credit & Debit Cards *(coming soon)*

---

Compatibility & Integration

Spoiler:

Compatible with any client that supports the OpenAI API.

Works with:

* Cursor
* Claude Code
* Codex CLI / App
* VS Code
* OpenCode
* OpenAI SDK
* Your own bots
* Scripts
* AI agents

Integration is extremely simple — just replace your `base_url` with the VibeLLM endpoint and provide your API key.

Endpoint
https://vibellm.store/v1

JavaScript Example

Spoiler:

codeblock Wrote:

 

Code:

javascript
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://vibellm.store/v1",
  apiKey: process.env.VIBELLM_API_KEY,
});

const res = await client.chat.completions.create({
  model: "gpt-5.4",
  messages: [
    { role: "user", content: "Refactor this component" }
  ],
});

console.log(res.choices[0].message.content);
[/quote]

---

What You Can Build

Spoiler:

* Vibe coding in Cursor, Codex, VS Code, and OpenCode
* Code generation and refactoring
* Bug fixing and debugging
* Working with large codebases
* Integrating AI into your own scripts
* Building Telegram bots
* Creating AI agents
* Workflow automation
* Testing different AI models through a single API

---

 Privacy & Security

Spoiler:

VibeLLM **does not store** the contents of your prompts, requests, or AI responses.

Only technical metadata required for billing is retained:

* Model used
* Token usage
* Request status

API keys are stored as secure hashes, all traffic is encrypted over HTTPS, and configurable usage limits are available for every API key.

---

FAQ

Spoiler:

 Does it work with Cursor and Codex?

Yes. It works with any client that allows you to specify an OpenAI-compatible endpoint.

Can I integrate it into my own project?

Yes. The endpoint works with SDKs, CLI tools, custom applications, bots, and AI agents.

Is there a subscription?

No. You only pay for the tokens you actually use.

Are my prompts stored?

No. Your prompts and AI responses are never stored.

[ro635]

Landing website doesn't mention that token usage increases up to 10x depending on model, but GLWS!

[noware]

23-07-2026, 09:28 PM ro635 Wrote:

Landing website doesn't mention that token usage increases up to 10x depending on model, but GLWS!

Its how it work, but its anyway up to 100x cheaper, than og api from claude or openai. Try our promotion and you will see ;3

[SignatureSpots]

Good luck

[NapoleonRefs]

GOOD LUCK WITH SALE !!!

[Alvinloso]

Really amazing tool, would love to continue using this!

[noware]

04-08-2026, 04:24 AM Alvinloso Wrote:

Really amazing tool, would love to continue using this!

Thank you!