Kilo
Summary
Vendor lock-in is quiet until it isn't — then you're either paying OpenAI's markup on every token or rewriting your workflow to swap models. Kilo Code is built around the premise that your agent and your model choice should be two separate decisions.
Kilo Code is an open-source (Apache 2.0) coding agent that runs inside VS Code, JetBrains IDEs, and the CLI, with cloud agent and Slack options on top. It ships five specialized modes — Code, Architect, Debug, Ask, and Custom — so you're not forcing a general-purpose chat model to plan a feature and then write it in the same session. The 500+ model catalog routes through Kilo Gateway at zero markup, which means your token bill reflects actual model pricing. That architecture holds up well for single-developer workflows and small teams. Where it gets complicated is at the org level: team-wide parallel workflows using isolated agent worktrees are a newer surface, and community reports suggest the tooling around coordinating those agents is still maturing.
Bottom line: Pick Kilo Code when your priority is model flexibility and cost transparency on a solo or small-team codebase — plan for additional orchestration work when you need coordinated multi-agent workflows across a larger engineering team.
Pricing Plans
SubscriptionLast verified 2 days ago- Price
- Free (extension); Kilo Pass $19–$199/month (credits); KiloClaw $55/month (cloud agent)
- Free Tier
- Free & Open Source - Free AI usage billed separately, open source VS Code/JetBrains/CLI extensions, agentic coding assistant, code completion & generation, works with any AI model
Free & Open Source
For every developer building with AI
- Free & Open Source
- Free AI usage billed separately
- Open source VS Code, JetBrains & CLI extensions
- Agentic coding assistant
- Code completion & generation
- Works with any AI model
Teams
Everything in Free, plus team management
- Team management
- Usage analytics & reporting
- AI adoption score
- Shared agent modes
- Centralized billing
- Shared BYOK
- Data privacy controls
- Priority support
Enterprise
Everything in Teams, plus security & compliance
- Custom SSO, audit logs & SLA
- Limit models and/or providers
- Shared Private Gateway
- BYOK
- Audit logs
- SSO, OIDC, & SCIM support
- SLA commitments
- Dedicated support channels
View full pricing on kilo.ai →
Pricing may have changed since last verified. Check the official site for current plans.
Community Performance Report Card
No community ratings yet. Be the first to rate this tool!
Community Benchmarks Community
Sign in to submit a benchmarkNo community benchmarks yet. Be the first to share a real-world data point.
Pros
Sign in to edit- Zero-markup model routing across 500+ providers, so your token cost reflects actual model pricing and switching models when costs spike is a config change rather than a platform migration.
- Five specialized agent modes (Code, Architect, Debug, Ask, Custom) split planning from execution, so you're not asking the same agent session to design an architecture and then write the implementation — context stays focused.
- Apache 2.0 core with self-hosted and air-gap deployment options, which means organizations with data residency requirements can run the agent without sending code to external infrastructure.
- BYOK support across 20+ providers according to the docs, so teams with existing enterprise model agreements don't pay a second time through the platform.
- KiloClaw managed cloud agents deploy without SSH, Docker, or yaml configuration, so teams that want 24/7 autonomous task execution don't need to maintain that infrastructure themselves.
Cons
Sign in to edit- Multi-agent parallel workflows using isolated worktrees are documented as a feature, but the tooling for coordinating agents across a shared codebase is less mature than the single-developer IDE flow — teams hitting this at scale report needing to build their own coordination layer on top.
- The five-mode system requires you to consciously switch contexts between planning and execution. Teams that want a single agent to move fluidly from architecture to implementation without manual mode switching find this model adds friction, and at that point tools with a more unified agent loop become the alternative they evaluate.
- KiloClaw (the managed cloud agent layer) is a paid-only feature, meaning teams that want the 'deploy in 60 seconds, no infrastructure' path are outside the free tier — the self-hosted option requires enough DevOps capacity to stand it up.
Community Reviews
Sign in to write a reviewNo reviews yet. Be the first to share your experience.
About
- Platforms
- VS Code, JetBrains (IntelliJ, PyCharm, WebStorm), CLI, Cloud Agents, Slack, Cursor, Windsurf
- API Available
- Yes
- Self-Hosted
- Yes
- Last Updated
- 2026-05-30T21:58:49.182Z
Best For
Who it's for
- Developers seeking model-agnostic flexibility without vendor lock-in
- Teams prioritizing transparent token-based pricing over subscription bundles
- Organizations requiring local-first or air-gap deployment options
- Workflows needing deep codebase understanding and multi-step autonomy
- Projects integrating external tools via Model Context Protocol
What it does well
- Multi-file refactoring and code generation from natural language
- Autonomous bug diagnosis and fix iteration across test suites
- Complex project planning and architecture design before implementation
- Team-based parallel coding with isolated agent worktrees
- Production code review with inline comments and diff analysis
Integrations
Discussion Community
Sign in to commentNo discussion yet. Sign in to start the conversation.
Compare Kilo
Spotted incorrect or missing data? Join our community of contributors.
Sign Up to ContributeCommunity Notes & Tips Community
Sign in to contributeBe the first to contribute. General notes, observations, gotchas, and tips from people who use this tool day-to-day.
Recommended skills for this tool
Auto-curated by the AIDiveForge recommendation matrix. These skills are predicted to enhance this tool based on category, capability, and domain signals.
-
Meeting Summary Template transform 32%
Turn a raw transcript into a decision-focused recap: outcomes, owners, deadlines, open threads.
Why: category partial · caps 0/0 · domain ops
-
Standup Note Synthesizer transform 32%
Merge individual standup bullets from multiple people into a single team digest with blockers surfaced to the top.
Why: category partial · caps 0/0 · domain ops
-
Runbook Skeleton post 32%
Produce a first-draft runbook from a postmortem — detection, diagnosis, mitigation, rollback — so the next incident has a template to follow.
Why: category partial · caps 0/0 · domain ops
-
OKR Draft Critiquer post 32%
Score draft OKRs against SMART criteria and the outcome-not-output rule, with suggested rewrites for each failing key result.
Why: category partial · caps 0/0 · domain ops
Frequently Asked Questions
- Is Kilo free?
- Kilo is a paid tool (Free (extension); Kilo Pass $19–$199/month (credits); KiloClaw $55/month (cloud agent)). A 14-day free trial is available.
- Is Kilo open source?
- No — Kilo is a closed-source tool. Source code is not publicly available.
- Does Kilo have an API?
- Yes. Kilo exposes a developer API. See the official documentation at https://kilo.ai for details.
- Can I self-host Kilo?
- Yes. Kilo supports self-hosting on your own infrastructure.
- When was Kilo released?
- Kilo was first released in 2025.
- What platforms does Kilo support?
- Kilo is available on: VS Code, JetBrains (IntelliJ, PyCharm, WebStorm), CLI, Cloud Agents, Slack, Cursor, Windsurf.
Hours Saved & ROI Stories Community
Sign in to contributeBe the first to contribute. Concrete time/cost savings, with context. e.g. "Cut my code review backlog from 4h to 45m per week."
Curated lists that include this category
Most AI coding agents either assume you’ll use their preferred model forever or charge a markup to support alternatives. Kilo Code separates the agent layer from the model layer entirely. The extension installs into VS Code or JetBrains, connects to 500+ models through Kilo Gateway, and operates in five distinct modes: Code for writing and refactoring, Architect for planning before implementation, Debug for tracing and fixing failures, Ask for codebase Q&A, and Custom for teams that want to define their own agent behavior. The agent can invoke file operations, run terminal commands, control a browser, and coordinate subagents — so it’s not just generating text, it’s executing steps.
The model-agnostic routing is the sharpest differentiator. The vendor states the gateway passes through model pricing at zero markup, which means switching from a frontier model to a cheaper alternative when a project’s budget tightens is a configuration change, not a migration. BYOK (bring your own key) is supported across 20+ providers according to the docs, so teams that already have enterprise agreements with a model provider can route through those instead of paying twice.
The self-hosted path is real — the core extension is Apache 2.0 licensed and the docs describe local-first and air-gap deployment options, which matters for organizations where code cannot leave the building. For teams that don’t want to manage infrastructure, KiloClaw is the vendor’s managed cloud agent layer: one-click deploy, auto-restart, and connections to Telegram, Discord, or Slack. The self-hosted and managed paths serve genuinely different buyers, and the boundary between them is clear enough to plan around.
Where Kilo Code shows its edges is in team-scale coordination. Parallel coding with isolated agent worktrees is listed as a supported use case, but this surface is newer than the single-developer IDE flow, and the tooling for managing multiple agents working on the same codebase simultaneously is less documented than the core modes. Teams pushing into that territory should expect to do integration work that the docs don’t fully cover yet.