$Cheapa API Docs

Kilo Code

Connect Kilo Code, the VS Code AI assistant, to Cheapa API for intelligent code generation and assistance.

What This Does

Kilo Code is an AI-powered coding assistant for VS Code that helps with code generation, explanations, debugging, and refactoring. By connecting Kilo Code to Cheapa API, you access a wide range of models through a single endpoint with consolidated billing.


Setup

Step 1: Install Kilo Code

  1. Open VS Code
  2. Go to the Extensions marketplace (Cmd+Shift+X or Ctrl+Shift+X)
  3. Search for Kilo Code
  4. Click Install

Step 2: Open Kilo Code Settings

  1. Click the Kilo Code icon in the VS Code sidebar
  2. Click the settings icon or navigate to Configuration
  3. Find the Providers section

Step 3: Configure Cheapa API

Set up Kilo Code with Cheapa API:

FieldValue
API ProviderOpenAI Compatible
Base URLhttps://api.cheapaapi.com/v1
API KeyYour Cheapa API API key
Modelkimi-k2.7-code or deepseek-v4-pro

Get your API key from cheapaapi.com/keys.

Step 4: Save and Verify

  1. Save the configuration
  2. Open the Kilo Code chat panel
  3. Send a test message to confirm the connection

Use CaseModelDescription
Code generation & refactoringkimi-k2.7-codeMoonshot's code-tuned K2.7 — best for code generation
Code generationdeepseek-v4-proBest for writing and modifying code
Code explanationskimi-k2.6Superior reasoning for complex logic
Large filesminimax-m3Extended context window
Quick assistanceglm-5.1Fast and economical

Troubleshooting

API key rejected

Model not recognized

  • Select OpenAI Compatible provider type
  • Use exact model slugs from the Models reference

No response from AI

Slow responses

  • Switch to glm-5.1 for faster responses
  • Check your network connection

Notes

  • Kilo Code supports inline code suggestions and chat-based interactions
  • All token consumption is billed to your Cheapa API account
  • Usage statistics appear in real-time on your Cheapa API dashboard
  • Consider using different models for different tasks to optimize cost and performance

On this page