AI CODING ASSISTANT

KCode
AI Coding Assistant by AstroLexis

100% local, runs on your GPU, your code never leaves your machine. A complete AI-powered coding toolkit with adaptive memory, 46 built-in tools, and zero cloud dependency.

KCODE
$ kcode
 _  _ ___ ___  ___  ___
| |/ / __/ _ \|   \| __|
| ' < (_| (_) | |) | _|
|_|\_\___\___/|___/|___|
Kulvex Code — AI Coding Assistant by Astrolexis
Hardware detected: RTX 5090 (32 GB VRAM)
Model: mnemo:mark6-max — Gemma 4 31B abliterated Q8_0 — 32GB+ GPUs
Inference server ready on port 10091
kcode > refactor the auth module to use JWT tokens
Reading src/auth/... Analyzing 12 files... Editing...

Everything you need to code with AI

A complete coding assistant that runs on your hardware. No cloud, no telemetry, no compromises.

Adaptive Memory

Remembers every successful interaction. Past solutions become future context — the AI gets better the more you use it.

Zero Configuration

Auto-detects your GPU, downloads the best model, starts the server. Just type kcode.

46 Built-in Tools

File ops, bash, grep, git, web access, LSP, multi-edit, cron, worktrees, planning, memory — a complete AI coding toolkit.

10-Layer Cognition

Identity, tools, metacognition, user model, world model, session narrative — not just a chatbot wrapper.

Enterprise Ready

Managed policies, audit logging, model restrictions, permission rules, SSRF protection, and org-level tool controls.

100% Local & Private

Your code never leaves your machine. No telemetry, no cloud. Protected directories, symlink resolution, and SSRF guards built in.

Powered by mnemo:mark6

Gemma 4 31B abliterated — local models that learn from you. Auto-selected based on your GPU.

ModelSizeMin VRAMBest for
mnemo:mark6-pico2.6 GB3 GBCompact 4B abliterated — 4GB GPUs or CPU
mnemo:mark6-nano6 GB12 GBFast dense 8B — 12GB GPUs
mnemo:mark6-mini13.5 GB16 GBGemma 4 dense 14B — 16GB GPUs
mnemo:mark6-mid18.6 GB24 GBGemma 4 31B Q4_K_M — 24GB GPUs
mnemo:mark6-max33 GB36 GBGemma 4 31B Q8_0 — best quality, 32GB+ GPUs
mnemo:mark6-31b33 GB36 GBFull Gemma 4 31B abliterated — flagship model

GGUF format — works on NVIDIA (CUDA), AMD (ROCm), Apple Silicon (Metal), and CPU

Download KCode v2.10.0

Single binary, no dependencies. The setup wizard handles everything automatically.

Free & open-source (AGPLv3). The setup wizard handles everything automatically.

Simple, transparent pricing

Start free, upgrade when you need more. No hidden fees, no cloud lock-in.

KCode Free

$0

Basic features, 1 session, community support

  • Local inference on your GPU
  • 17 core tools
  • 1 active session
  • Community support
  • Basic memory
MOST POPULAR

KCode Pro

$19/mo

All features, 8 swarm agents, unlimited sessions, deploy automation

  • Everything in Free
  • 46 built-in tools
  • 8 swarm agents
  • Unlimited sessions
  • Deploy automation
  • Priority support
  • Advanced memory + RAG

KCode Solo

$99 one-time

Everything in Pro, permanent license, no subscription

  • Everything in Pro
  • Permanent license
  • No subscription ever
  • Lifetime updates
  • Priority support

Volume licensing: contact sales

Your code. Your AI. Your machine.

No cloud. No telemetry. No subscription required. Download KCode and start coding with AI that runs 100% on your hardware.