Max

CLI wrappers that route Claude Code / Codex through multiple API-compatible endpoints:

• max - MiniMax API — routes Claude Code via Anthropic-compatible API
• glm - Z.ai GLM — routes Claude Code via Anthropic-compatible API
• klaude - Moonshot AI Kimi — routes Claude Code via Anthropic-compatible API
• kimix - Kimi for Coding (k3-256k) — routes Codex CLI via a local proxy that translates Responses API ↔ Chat Completions

Use Claude Code or Codex with your preferred provider without modifying configuration each time.

Requirements

• Claude Code CLI installed and in your PATH (for max, glm, klaude)
• Codex CLI installed and in your PATH (for kimix)
• Python 3 with flask and requests (for kimix proxy — installed automatically)
• API key(s) for your chosen provider(s)

Setup

1. Install the binaries

curl -fsSL https://raw.githubusercontent.com/UnlikeOtherAI/max/main/install.sh | bash

Or manually:

# Install all four
curl -fsSL https://raw.githubusercontent.com/UnlikeOtherAI/max/main/max -o ~/bin/max
curl -fsSL https://raw.githubusercontent.com/UnlikeOtherAI/max/main/glm -o ~/bin/glm
curl -fsSL https://raw.githubusercontent.com/UnlikeOtherAI/max/main/klaude -o ~/bin/klaude
curl -fsSL https://raw.githubusercontent.com/UnlikeOtherAI/max/main/kimix -o ~/bin/kimix
curl -fsSL https://raw.githubusercontent.com/UnlikeOtherAI/max/main/kimix-proxy -o ~/bin/kimix-proxy
curl -fsSL https://raw.githubusercontent.com/UnlikeOtherAI/max/main/k3-catalog.json -o ~/bin/k3-catalog.json
chmod +x ~/bin/max ~/bin/glm ~/bin/klaude ~/bin/kimix ~/bin/kimix-proxy

k3-catalog.json must sit next to the kimix script — that's where it picks up the model list.

Ensure ~/bin is in your $PATH.

Additional setup for kimix

kimix starts a local HTTP proxy (kimix-proxy) that translates the Codex Responses API to Kimi's Chat Completions API. This lets the upstream Codex binary run unmodified.

One-time Python setup:

python3 -m venv ~/.kimix/venv
~/.kimix/venv/bin/pip install flask requests

Save your API key:

mkdir -p ~/.kimix && echo "your_kimi_coding_api_key" > ~/.kimix/token && chmod 600 ~/.kimix/token

That's it — kimix handles everything else on first run.

2. Provide your API key(s)

• MiniMax: export MINIMAX_API_KEY=your_key_here
• Kimi (klaude): export KIMI_CODING_API_KEY=your_key_here
• Kimi (kimix): export KIMIX_API_KEY=your_key_here or save to ~/.kimix/token
• GLM: export ZAI_CODING_PLAN_API_KEY=your_key_here

Or use on first run (interactive):

max "your prompt"    # MiniMax — prompts for key if not set, saves to ~/.max/token

Note: glm and klaude require their respective environment variables to be set in your shell.

Usage

Each CLI tool passes all arguments directly to Claude Code or Codex:

max "refactor the auth module"
max --print "what changed in the last commit"
max --input "file.txt" "explain this"

glm "refactor the auth module"
glm --print "what changed in the last commit"
glm --input "file.txt" "explain this"

klaude "refactor the auth module"
klaude --print "what changed in the last commit"
klaude --input "file.txt" "explain this"

kimix "refactor the auth module"
kimix "what changed in the last commit"

Choosing a Kimi model

kimix defaults to k3-256k — the same K3 model with a 256k context window. Plain k3 is the full 1M-context variant and costs substantially more, so it's opt-in:

kimix -m k3 "trace this bug across the whole monorepo"

Or set it for a shell session:

export KIMIX_MODEL=k3

kimi-for-coding and kimi-for-coding-highspeed (K2.7, 256k) are also selectable the same way.

How it works

Each CLI tool sets environment variables before spawning Claude Code:

max (MiniMax):

ANTHROPIC_BASE_URL=https://api.minimax.io/anthropic
ANTHROPIC_AUTH_TOKEN=<your MINIMAX_API_KEY>

glm (Z.ai):

ANTHROPIC_BASE_URL=https://api.z.ai/api/anthropic
ANTHROPIC_AUTH_TOKEN=<your ZAI_CODING_PLAN_API_KEY>
ANTHROPIC_MODEL=glm-4.6
ANTHROPIC_SMALL_FAST_MODEL=glm-4.5-air

klaude (Kimi/Moonshot AI):

ANTHROPIC_BASE_URL=https://api.kimi.com/coding/
ANTHROPIC_AUTH_TOKEN=<your KIMI_CODING_API_KEY>
ANTHROPIC_MODEL=kimi-for-coding

kimix (Kimi K3 via Codex + proxy):

KIMIX_API_KEY=<from $KIMIX_API_KEY or ~/.kimix/token>
KIMIX_MODEL=k3-256k               (default; `k3` = 1M context, much pricier)
CODEX_HOME=~/.kimix/codex-proxy   (isolated config: kimi provider, k3-256k model)

kimix starts kimix-proxy on http://127.0.0.1:8787 — a local Flask server that translates Codex's OpenAI Responses API requests to Kimi Chat Completions requests, and streams responses back in Responses API SSE format. Codex's config.toml points to the proxy as a custom provider with wire_api = "responses". The proxy is shared across sessions and stays running in the background.

Everything else — conversation history, tool use, etc. — is handled by the respective CLI as usual.

License

MIT