Kilo Code
» AI Agents » Ready-to-Use » Kilo Code URL Share on
Open-source, model-agnostic AI coding agent that works as a VS Code/JetBrains extension and a standalone CLI, letting developers pick from 500+ LLMs and switch models mid-task.
Key Features
Multi-Agent Single Agent Human-in-the-Loop Streaming Async Support Type Safe Short-Term Memory Plugin System Custom Tools MCP Protocol Code Execution Web Browsing File System Access CLI Self-Hosted Cloud Hosted
Community Feedback
Strengths
- No vendor lock-in: BYOK access to 500+ models at provider pricing
- Fork lineage from Roo Code/Cline gives it mature agent-mode UX (Code/Plan/Ask/Debug/Review)
- Ships in VS Code, JetBrains, CLI, and a cloud agent, covering most developer workflows
Weaknesses
- Young, fast-moving codebase with very high release velocity, which can mean frequent breaking changes
- Now owned by Anaconda, Inc. (acquired Jul 2026), raising questions about long-term open-source direction
- Requires a provider API key/account to access most models; no bundled free inference
Kilo Code Details
| Organization | Kilo Code (Anaconda, Inc.) |
| Organization Type | Company |
| Funding | Acquired |
| Category | Ready-to-Use |
| Subcategory | Code generation |
| Deployment | Ide Extension Cli |
| Primary Language | TypeScript |
| Runtime | VS Code / JetBrains extension host; Node.js CLI (fork of OpenCode) |
| License | MIT |
| Commercial Use | Unrestricted |
| Install Command | npm install -g @kilocode/cli |
| GitHub Stars | 26,473 |
| GitHub Forks | 2,962 |
| Open Issues | 705 |
| Contributors | 1,164 |
| Last Commit | 2026-07-23 |
| First Release | 2025-03-10 |
| Release Cadence | Very frequent (near-daily builds) |
| Maturity | Stable |
| Pricing Model | Freemium |
| Free Tier | Open-source extension/CLI is free; users bring their own API key or pay providers directly at cost (no markup) |
| Self-Hosted Free | Yes |
| Cost Model | free + pay LLM provider's own rate (zero markup) |
| Community Size | Large (26.4K stars, 3M+ reported users) |
| Community Activity | Very active |
| Sentiment | Positive |
| GPU Required | No |
| Research Date | 2026-07-23 |
| LLM Providers | Anthropic, OpenAI, Google Gemini, 500+ models via OpenRouter and direct provider keys |
| API Keys Required | LLM provider API key (or Kilo account for provider-priced access) |
Use Cases
- In-editor code generation and multi-file refactoring from natural-language prompts
- Autonomous CI/CD tasks via `kilo run --auto` (e.g., running tests and fixing failures unattended)
- Architecture planning and implementation-plan drafting before code is written (Plan mode)
- Automated pull request code review across performance, security, style, and test coverage
- Terminal and browser automation driven by the agent
When to Use
Best for: Teams that want an open-source, editor-native coding agent with freedom to choose and switch LLM providers without markup
Avoid when: You need a single vendor-supported, fully closed enterprise IDE with built-in model hosting and no external API key management
Original data from HuggingFace, Arena and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a