Coding assistants
Kilo
Made by Kilo
An open-source coding assistant that runs in your editor and uses any model you plug in.
What it is
Kilo is the wrapper, not the brain. It sits in VS Code, JetBrains, the terminal or the cloud, reads your project, and applies changes — using whichever model you point it at, at that provider's list price with no markup.
For someone who doesn't write code, this is a useful window into how a site like this one gets built: you describe a change in plain English and something else edits the files.
Free for individuals. Team features are the paid part, and the model cost is separate — though its free model routing means you can start at nothing.
Published price
- IndividualOpen source, free forever$0
- Model useOr $0 with the free model routingProvider cost, zero markup
- Teams14-day trial$15/user/month
- EnterpriseCustom
Source: Kilo's pricing page · Last checked 17 September 2026
Benchmarks
The published figures for Kilo, copied exactly as their sources give them and checked 17 September 2026. Each one names the test, who ran it and when, and says when the number comes from the maker rather than an independent tester.
- KiloBench (Terminal-Bench 2.0, run in Kilo's harness)76.2% at $87.41 per attempt — 1st
share of real command-line tasks completed unaided, with the API cost of trying · tested: GPT-5.6 Sol driven by Kilo
Kilo's own KiloBench board · checked September 2026 · maker's own figures
- KiloBench (Terminal-Bench 2.0, run in Kilo's harness)75.3% at $112.27 per attempt — 3rd
share of real command-line tasks completed unaided, with the API cost of trying · tested: Gemini 3.8 Flash driven by Kilo
Kilo's own KiloBench board · checked September 2026 · maker's own figures
- KiloBench (Terminal-Bench 2.0, run in Kilo's harness)73.0% at $33.83 per attempt — 5th
share of real command-line tasks completed unaided, with the API cost of trying · tested: Grok 4.6 driven by Kilo
Kilo's own KiloBench board · checked September 2026 · maker's own figures
Worth reading next to the row above: three points behind the leader for a third of the money.
Kilo now publishes its own board — KiloBench — which runs Terminal-Bench 2.0 through Kilo's actual agent harness and reports the cost of each attempt alongside the pass rate. These are Kilo's own figures, not an independent leaderboard, and each one belongs to the model plugged in rather than to Kilo itself. The 88% SWE-bench figure floating around one comparison blog still has no primary source, so it stays out.
Figures copied from the sources above, checked 17 September 2026.
Where to read today's numbers
Scores move, so these are the boards that keep them current.
- SWE-bench leaderboardIndependent
Measures whether an agent can fix real GitHub issues. The standard reference for coding agents.
- Artificial AnalysisIndependent
Independent testing of quality, speed and price per million tokens across hosted models.
How it performs, in my experience
As good as whichever model you plug into it. Works in VS Code, JetBrains, the terminal and the cloud.
Free value
4 / 5
How much you get without paying.
Performance
4 / 5
How well it does its main job.
Range of uses
4 / 5
How many different jobs it suits.
Free tool, but you pay for whichever model you plug into it. These scores are my own judgment, not a measurement — the benchmark links above are the independent version.
Prompting it from this site
Can't be run from here
A coding assistant that edits files in your editor, not a model you send one prompt to.
Use it for
- Editing a project without paying a subscription
- Keeping code private by running a local model
Running it yourself
The software is local, the model isn't
Kilo is open-source software that runs inside your editor on your machine — but the thinking is done by whichever model you point it at, which usually means a paid key or a local model.
What you need first
- Visual Studio Code (or a compatible editor)
- Node.js 20+ only if you build it from source
- Either an API key for a hosted model, or a local model already running
Step by step
1.Install the extension from the editor's marketplace — search for Kilo Code
2.Or build it yourself
git clone https://github.com/Kilo-Org/kilocode && cd kilocode && npm install && npm run build3.Point it at a local model instead of a paid API
ollama run qwen2.5-coder # then choose the Ollama provider in Kilo's settings
Running it against a local model is free but noticeably slower and weaker at long tasks than the paid frontier models.