Tune Models

LLM models middleware for Tune - chat with AI directly in your text editor.

This middleware connects popular AI models like OpenAI, Anthropic, and others to your Tune setup.

Available Models

 system: @gpt-4.1-mini

 user: 
 what is 2+2?

 assistant: 
 2+2 equals 4.

 system: @claude-3-sonnet  

 user: 
 explain quantum computing briefly

 assistant: 
 Quantum computing uses quantum bits that can exist in multiple states...

Supports:

• OpenAI: gpt-4, gpt-4.1-mini, gpt-3.5-turbo, etc.
• Anthropic: claude-3-5-sonnet, claude-3-haiku, claude-3-opus, etc.
• OpenRouter: Access to hundreds of models from various providers
• Google Gemini: gemini-pro, gemini-1.5-pro, gemini-flash, etc.
• Mistral: mistral-large, mistral-medium, mistral-small, etc.
• Groq: llama3-70b-8192, mixtral-8x7b-32768, gemma-7b-it, etc.
• Ollama whatever models are pulled in your ollama server

It fetches the list of latest models from provider's api.

Setup for Text Editor

Install globally in your ~/.tune folder:

cd ~/.tune
npm install tune-models

Add to ~/.tune/default.ctx.js:

const models = require('tune-models')

module.exports = [
    ...
    models({
        default: "gpt-4.1",
        ...
    })
    ...
]

Set your API keys in environment or ~/.tune/.env:

OPENAI_KEY=sk-...
ANTHROPIC_KEY=sk-ant-...
OPENROUTER_KEY=sk-or-...
GEMINI_KEY=AI...
MISTRAL_KEY=...
GROQ_KEY=gsk_...
OLLAMA_URL="http://localhost:11434"

Setup for JavaScript Project

npm install tune-models tune-sdk

const tune = require('tune-sdk')
const { openai, anthropic, groq, mistral, geimini, openrouter, ollama } = require('tune-models')

const ctx = tune.makeContext(
    { OPENAI_KEY: process.env.OPENAI_KEY }, 
    openai({ expose: ["gpt-4.1", "gpt-4.1-mini"], cache: false }), 
    anthropic({ apiKey: 'sk-ant-...' }))

const result = await ctx.text2run(`
 system: @gpt-4.1-mini
 user: hello
`)

Configuration Options

All Providers

models({
  // Set default model when no @model specified
  default: "gpt-4.1-mini",
  
  // Create aliases for models  
  alias: { "gpt": "gpt-4.1-mini" },
  
  // Only expose specific models
  expose: ["gpt-4.1-mini", "claude-3-sonnet"],
  
  // Mount under prefix: @models/gpt-4.1-mini
  mount: "models",
  
  // Cache model lists on disk (default: true)
  cache: true,
  cacheTtl: 3600000, // 1 hour
  
  // API keys for all providers
  apiKeys: {
    openai: process.env.OPENAI_KEY,
    anthropic: process.env.ANTHROPIC_KEY,
    ollama: process.env.OLLAMA_URL
  }
})

Provider-Specific

openai({ 
  default: "gpt-4.1-mini",
  apiKey: "sk-...",
  mount: "openai",
  expose: ["gpt-4.1-mini"]
  cache: true,
})