How to Use

Got your free credits but not sure where to start? Pick your favorite tool below — three steps and you're good.

📋 Two Things You Need

🔑 API Key

Create one at Settings → API Keys. Looks like sk-...

🌐 Base URL

Most tools use:
https://llmrpc.com/v1

Some tools auto-append paths — use:
https://llmrpc.com

Each tool section below specifies the correct format.

💡 Treat your API key like a password. It's only shown once when created — copy and save it somewhere safe now.

🖥️ Using in VSCode

Install an AI extension and get inline chat, autocomplete, and code review right in your editor.

🔷Continue (Recommended • Free & Open Source)

Lightweight AI coding assistant with autocomplete + chat. 2M+ installs. Fast and perfect for daily use.

  1. Search Continue in the VSCode Extensions marketplace and install
  2. Open the config file: ~/.continue/config.json
  3. Paste the config below (replace YOUR_API_KEY with your actual key)
  4. Cmd/Ctrl + Shift + P → type Continue: Focus → start coding
Show config
{
  "models": [
    {
      "title": "DeepSeek V4 Pro",
      "provider": "openai",
      "model": "deepseek-v4-pro",
      "apiBase": "https://llmrpc.com/v1",
      "apiKey": "YOUR_API_KEY"
    },
    {
      "title": "Kimi K2.6",
      "provider": "openai",
      "model": "kimi-k2.6",
      "apiBase": "https://llmrpc.com/v1",
      "apiKey": "YOUR_API_KEY"
    }
  ],
  "tabAutocompleteModel": {
    "title": "DeepSeek V4 Flash",
    "provider": "openai",
    "model": "deepseek-v4-flash",
    "apiBase": "https://llmrpc.com/v1",
    "apiKey": "YOUR_API_KEY"
  }
}
🟠Cline (Most Powerful • Open Source)

Autonomous AI coding agent that reads files, writes code, and runs terminal commands. More powerful but uses more tokens.

  1. Search Cline in the VSCode Extensions marketplace and install
  2. Open Cline panel (robot icon in sidebar) → click ⚙️ settings
  3. Set API Provider to OpenAI Compatible
  4. Fill in Base URL, API Key, and Model ID (see config below)
  5. Return to the chat panel and start building
Show config
# In Cline settings panel:
Base URL: https://llmrpc.com/v1
API Key:  YOUR_API_KEY
Model ID: deepseek-v4-pro
💡 Continue for everyday autocomplete & chat; Cline when you want the AI to directly edit your codebase. Start with Continue, upgrade to Cline when needed.

💬 ChatBox (Desktop Chat App)

ChatBox is a cross-platform desktop & mobile AI client — perfect if you just want a ChatGPT-style chat interface without writing any code.

📦ChatBox

Available on Windows, macOS, Linux, iOS, and Android. Clean UI, easy to set up.

  1. Download from chatboxai.app
  2. Settings → Model Provider → select OpenAI API
  3. API Host: https://llmrpc.com (leave API Path as default — ChatBox auto-appends /v1/chat/completions)
  4. API Key: your sk-... key
  5. Model: deepseek-v4-pro (or pick any from the Models page)
  6. Save and start chatting 🎉

🤖 LobeChat

LobeChat is an open-source, beautifully designed AI chat framework with plugins, knowledge base, and TTS support.

🎨LobeChat

Beautiful web UI with plugin system, custom agents, and knowledge base. Self-host or use the cloud version.

  1. Visit lobechat.com or the community edition
  2. Settings → Language Model → OpenAI → Enable
  3. API Proxy URL: https://llmrpc.com/v1
  4. API Key: your sk-...
  5. Custom model list: deepseek-v4-pro,deepseek-r1,kimi-k2.6,qwen3.6-27b,glm-5.2
  6. Start chatting

✨ NextChat (ChatGPT Next Web)

NextChat is the original popular ChatGPT open-source frontend — extremely lightweight and easy to deploy.

⚡NextChat

Lightweight web chat. Deploy with one Docker command or one-click Vercel deploy.

  1. Open NextChat web app or your self-hosted instance
  2. Bottom-left ⚙️ → expand Advanced Settings
  3. API Endpoint: https://llmrpc.com (do NOT include /v1 — NextChat auto-appends it)
  4. API Key: your sk-... key
  5. Custom model name: deepseek-v4-pro
  6. Save → start using

🌐 Open WebUI

Open WebUI is the most feature-rich self-hosted AI platform — great for teams and power users.

🐳Open WebUI

Most complete self-hosted option: RAG knowledge base, multi-user management, web search, model management. Start with one Docker command.

  1. docker run -d -p 3000:8080 --name open-webui ghcr.io/open-webui/open-webui:main
  2. Open http://localhost:3000, create an admin account
  3. Settings → Admin Settings → External Connections → Add OpenAI API
  4. URL: https://llmrpc.com/v1 | Key: your API key
  5. Back to chat — select any LLMRpc model from the dropdown at the top-left

⌨️ Call the API from Your Code

Fully OpenAI SDK-compatible. Drop-in replacement — just change the base URL and API key.

cURL

curl https://llmrpc.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "deepseek-v4-flash",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Python

from openai import OpenAI

client = OpenAI(
    base_url="https://llmrpc.com/v1",
    api_key="YOUR_API_KEY"
)

response = client.chat.completions.create(
    model="deepseek-v4-flash",
    messages=[{"role": "user", "content": "Hello!"}]
)

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

Node.js

import OpenAI from 'openai'

const client = new OpenAI({
  baseURL: 'https://llmrpc.com/v1',
  apiKey: process.env.LLMRPC_API_KEY,
})

const response = await client.chat.completions.create({
  model: 'deepseek-v4-flash',
  messages: [{ role: 'user', content: 'Hello!' }],
})

console.log(response.choices[0].message.content)
💡 Any OpenAI-compatible library or tool works. Just swap base_url to https://llmrpc.com/v1.

📊 Popular Models Quick Reference

deepseek-v4-pro
DeepSeek V4 Pro
Flagship
deepseek-r1
DeepSeek R1
Deep Reasoning
deepseek-v4-flash
DeepSeek V4 Flash
Ultra Fast
qwen3.6-27b
Qwen3.6 27B
Latest Qwen
qwen3.5-122b
Qwen3.5 122B
Massive Capacity
kimi-k2.6
Kimi K2.6
All-round
glm-5.1
GLM-5.1
Solid Mid-tier
glm-5.2
GLM-5.2
Best Value

View all 60+ models & pricing →

❓ FAQ

"Invalid API Key" error?

Make sure your key starts with sk-. If you lost it, go to Settings → API Keys, delete the old one, and create a new one.

"Insufficient credits" error?

You're out of credits. Head to Billing to top up, or check if you still have unclaimed free credits.

Why no GPT or Claude models?

LLMRpc aggregates open-source and domestic models (DeepSeek, Qwen, GLM, Kimi, etc.). We do not proxy OpenAI or Anthropic proprietary models.

Do my credits expire?

Free credits never expire. Subscription quotas refresh monthly.

Can I use the same key across multiple tools?

Yes! One key works everywhere — VSCode, ChatBox, your own projects — all share the same balance.

Other tools not listed here?

Anything that supports OpenAI-compatible API works. Look for "Custom Endpoint", "OpenAI Compatible", or "API Base URL" in settings. Try https://llmrpc.com/v1 first. If you get a 404 or redirect, switch to https://llmrpc.com instead — some tools auto-append the /v1 path.

🎉 Not signed up yet?

New users get 100,000 free credits, plus 100,000 every month (claim manually). No credit card required.

Sign Up Free →