Tools/getkimchi/kimchi

kimchi

Terminal coding agent powered by Kimchi's multi-model orchestration

2.3k ★+5/wkemergingTypeScriptApache License 2.0 →trending

The Lens

By Erik Loyd, SaaS CEO and former COO/CFO of an AWS Premier Partner.

Updated Oct 2026

Kimchi is a terminal coding agent from the team behind Cast AI, built on the pi agent SDK. Its twist is multi-model routing: an orchestrator splits a job into explore, plan, build and review steps and hands each to a different model, light ones for reading code, heavier ones for hard builds. The CLI is free and Apache-2.0.

Install is one Homebrew or curl line, then kimchi setup. Every default points at Kimchi's hosted models (Kimi, MiniMax, Nemotron), and the docs assume an account. You don't strictly need one: it auto-detects local Ollama models and takes other providers' keys, but then you configure the role pools yourself.

The paid side is hosted inference. Coder is $20/month with $25 of credits, max 3 seats. Teams is $35/seat/month with budgets and RBAC. Claude or GPT routed through Kimchi adds $0.20/$0.30 per million tokens on top of provider pricing. Solo: free with your own keys, or Coder. Small teams: Coder, then Teams past 3 seats. Larger orgs: Enterprise for VPC deploys.

The catch: the CLI is a funnel for that inference business, and telemetry is on by default (usage metadata, not prompts) until you set KIMCHI_TELEMETRY_ENABLED=0.

Free vs Self-Hosted vs Paid

open core

Free: The kimchi CLI (Apache-2.0): multi-model orchestration, Ferment cross-session project mode, LSP code intelligence, MCP servers (with one-shot migration from Claude Code, OpenCode and Cursor), Bash hooks and ACP for IDEs. Kimchi's Open Source plan is bring-your-own-inference: "Kimchi's managed API, an OpenAI-compatible endpoint, or any LLM provider." Local Ollama models are discovered automatically.

Self-hosted: A standalone binary via Homebrew, a curl install script or PowerShell on macOS, Linux and Windows. Telemetry (OTLP usage metadata plus a PostHog device ID; prompts and file contents excluded, file paths hashed) is on by default; turn it off with KIMCHI_TELEMETRY_ENABLED=0 or telemetry.enabled: false in ~/.config/kimchi/config.json. Self-hosting the Kimchi platform itself in your VPC is listed for Teams and Enterprise.

Paid: Coder $20/month (includes $25 monthly credits, max 3 seats, 1 team). Teams $35/seat/month (includes $45 monthly credits, max 150 seats, Kimchi Studio, team budgets and RBAC, cloud marketplace billing). Enterprise is custom (unlimited seats, self-hosted VPC option, 24x7 dedicated CSM). Code Review add-on $15/user/month. Hosted models bill per million tokens, from DeepSeek V4 Flash at $0.14 input / $0.28 output to Kimi K3 at $3.00 / $15.00. External providers routed through Kimchi carry a flat $0.20 input / $0.30 output per million surcharge on top of the provider's own price.

The CLI is free with your own keys or local Ollama models. Kimchi's hosted models start at $20/month on Coder (with $25 of credits, 3 seats max), and teams pay $35/seat/month.

What to do by team size

Solo
free CLI with your own keys or Ollama; Coder at $20/month if you want Kimchi's hosted models and routing
Small team
Coder covers up to 3 seats; past that, Teams at $35/seat/month
Medium team
Teams at $35/seat/month for team budgets and RBAC, up to 150 seats
Large team
Enterprise for unlimited seats, VPC self-hosting and volume pricing
Self-hosting ops:trivial
View pricing page →

Get tools like this every Wednesday

One featured tool, three on the radar. No fluff.

Score
74/100 · B+
Adoption17/30
Maintenance25/25
Community7/20
License15/15
Analysis10/10

A low score is not a verdict on quality. Young and niche tools start low by design. How we calculate scores

Trust Signals

Organization account (7 public repos)

License: Apache License 2.0

Use freely. Patent grant included.

Commercial use: ✓ Yes

About

Owner
Kimchi (Organization)
Stars
2,276
Forks
150

Explore Further

More tools in the directory