Kilocode logo

Kilocode

Free

Kilo is the all-in-one agentic engineering platform. Build, ship, and iterate faster with the most popular open source coding agent.

FreeFree tier
Inputs: text, codeOutputs: code, text
Type
Open Source

About Kilocode

Kilo is an open-source AI coding agent that operates across IDEs (VS Code, JetBrains), CLI, and cloud environments, enabling developers to build, ship, and iterate faster. It offers multiple agent modes (Code, Architect, Debug, Custom) and supports 500+ AI models with zero markup through its Gateway. Key capabilities include code generation, autocomplete, voice prompting, automated code reviews, cloud agents for long-running tasks, one-click deployment, and session handoff across platforms. Kilo is MIT-licensed, fully inspectable, and can be self-hosted. Pricing is transparent: the open-source agent is free; AI inference is billed separately via pay-as-you-go or subscription (Kilo Pass), with a free tier available for local models or BYOK.

Key Features

Use Kilo Everywhere: VS Code, JetBrains, CLI, Cloud Agents, Slack
Agent Modes: Code, Architect, Debug, Custom, and Hosted (KiloClaw)
Auto Model routing: picks the best model based on task and budget (Efficient, Frontier, Balanced, Free)
Cloud Agents: run long-running, resource-intensive tasks remotely
Automated Code Review: AI agent analyzes PRs for bugs, style, and improvements
One-click deployment: deploy apps directly from the IDE with intelligent configuration detection
Agent Manager: run multiple agents in the IDE with embedded terminals and status badges
Sessions: seamless context handoff across devices and platforms
Voice Prompting: natural language voice commands in the IDE
Autocomplete in JetBrains IDEs: context-aware code completions

Pros & Cons

Pros
  • Fully open source with MIT license – no vendor lock-in
  • Works across multiple platforms: VS Code, JetBrains, CLI, cloud, and Slack
  • Transparent pricing with zero markup on AI inference
  • Auto Model routing optimizes for cost and capability automatically
  • Supports bring-your-own-key and local models (Ollama, LM Studio)
  • Multiple specialized agent modes (Code, Architect, Debug, Custom) for different workflows
  • Extensible via Skills and Gateways
Cons
  • AI inference is billed separately; free tier limited to specific free models or BYOK
  • Some advanced features (Teams, Enterprise, KiloClaw) require paid plans
  • Learning curve for configuring agent modes and Auto Model routing
  • Mobile apps (iOS/Android) currently limited to chat and session monitoring

Best For

Accelerate feature development with AI-assisted coding and architectural planningAutomate code reviews and ensure code quality across pull requestsRun long-running, complex tasks in the cloud without consuming local resourcesDeploy applications with one-click configuration detection and setupDebug and fix issues with an agent that reads errors and traces problemsManage multiple AI agents and sessions across IDEs, CLI, and mobileImplement scheduled tasks and cron jobs via always-on cloud agentsOnboard teams to AI-assisted development with centralized billing and analytics

FAQ

Is Kilo Code really free?
Yes, the Kilo Code open-source agent is free to use. AI inference costs are separate; you can use free models via Auto Free, bring your own keys, or pay as you go.
What is AI inference and why is it priced separately?
AI inference is the cost of running AI models (e.g., GPT-4, Claude). Kilo charges no markup on inference—you only pay the provider rates when using Kilo Gateway or Kilo Pass subscriptions.
What's the difference between Kilo Gateway and Kilo Pass?
Kilo Gateway is pay-as-you-go hosted AI inference with no subscription (pay exact provider rates). Kilo Pass is a monthly credit subscription that gives up to 50% bonus credits on top of your payment.
How does Auto Model affect AI inference cost?
Auto Model routes your tasks to the most appropriate model based on your selected tier (Free, Balanced, Frontier, Efficient). It automatically balances cost and capability, helping you save on inference spend.
What do Teams and Enterprise plans add?
Teams ($15/user/month) adds usage analytics, centralized billing, team management, shared BYOK, and data privacy controls. Enterprise adds SSO, audit logs, SLA, custom security, and dedicated support.
Do I need a subscription to start?
No. You can start with the Free open-source plan and choose to pay for AI inference later via pay-as-you-go or bring your own keys.