langtrain 0.1.11 → 0.1.13

package/README.md

@@ -1,100 +1,69 @@
-<div align="center">
+<p align="center"><code>npm i -g langtrain</code></p>
+<p align="center"><strong>Langtrain CLI</strong> is a unified AI engineering platform that runs locally on your computer and connects to the Langtrain Cloud.
+<p align="center">
   <picture>
     <source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/langtrain-ai/langtrain-sdk/main/public/langtrain-white.svg">
     <source media="(prefers-color-scheme: light)" srcset="https://raw.githubusercontent.com/langtrain-ai/langtrain-sdk/main/public/langtrain-black.svg">
-    <img alt="Langtrain Logo" src="https://raw.githubusercontent.com/langtrain-ai/langtrain-sdk/main/public/langtrain-black.svg" width="250">
+    <img alt="Langtrain Logo" src="https://raw.githubusercontent.com/langtrain-ai/langtrain-sdk/main/public/langtrain-black.svg" width="60%">
   </picture>
-  <h3>Langtrain SDK</h3>
-  <p>
-    The unified intelligence layer for JavaScript applications. <br/>
-    Combine computer vision and LLM fine-tuning in a single, high-performance SDK.
-  </p>
-  
-  <p>
-    <a href="https://www.npmjs.com/package/langtrain"><img src="https://img.shields.io/npm/v/langtrain?style=flat-square&labelColor=231f20&color=000000" alt="npm version" /></a>
-    <a href="https://langtrain.ai"><img src="https://img.shields.io/badge/website-langtrain.ai-000000?style=flat-square&labelColor=231f20" alt="website" /></a>
-    <a href="https://docs.langtrain.ai"><img src="https://img.shields.io/badge/docs-documentation-000000?style=flat-square&labelColor=231f20" alt="documentation" /></a>
-    <a href="https://github.com/langtrain-ai/langtrain-sdk/blob/main/LICENSE"><img src="https://img.shields.io/npm/l/langtrain?style=flat-square&labelColor=231f20&color=000000" alt="license" /></a>
-  </p>
-
-  <br/>
-
-  <pre>npm install langtrain</pre>
-
-  <br/>
-</div>
-
-## Overview
-
-Langtrain brings the power of **Langvision** (Computer Vision) and **Langtune** (LLM Optimization) into your JavaScript/TypeScript workflow. It is designed to be the only SDK you need to build intelligent, multimodal AI applications.
-
-## Key Features
-
-| Feature | Description |
-| :--- | :--- |
-| **👁️ Computer Vision** | Advanced image analysis, object detection, and visual reasoning powered by `Langvision`. |
-| **🧠 LLM Optimization** | Fine-tune and optimize language models effortlessly with `Langtune`. |
-| **📦 Unified & Bundled** | A single dependency for your entire AI stack. Zero configuration required. |
-| **⚡ Type-Safe** | Built with TypeScript for a robust, developer-friendly experience. |
-
-## Quick Start
-
-```typescript
-import { Langvision, Langtune } from 'langtrain';
-
-// 1. Initialize Clients
-const vision = new Langvision({ apiKey: process.env.LANGVISION_API_KEY });
-const tune = new Langtune({ apiKey: process.env.LANGTUNE_API_KEY });
-
-// 2. Build Intelligence
-async function analyzeAndTune() {
-  try {
-    // Analyze visual user context
-    const context = await vision.analyze({
-      image: 'https://example.com/dashboard.jpg',
-      features: ['layout', 'text'] 
-    });
-
-    console.log('Visual Context:', context);
-
-    // Generate optimized response based on visual context
-    const response = await tune.generate({
-      model: 'gpt-4-vision-optimized',
-      prompt: `Explain this dashboard layout: ${context.description}`
-    });
-
-    console.log('AI Response:', response);
-  } catch (error) {
-    console.error('Error processing AI request:', error);
-  }
-}
-
-analyzeAndTune();
+</p>
+<br/>
+<p align="center">
+If you want Langtrain in your code editor, <a href="https://langtrain.ai/docs/vscode">install the VS Code Extension.</a>
+<br/>If you are looking for the <em>web-based platform</em>, <strong>Langtrain Web</strong>, go to <a href="https://langtrain.ai">langtrain.ai</a>.
+</p>
+
+---
+
+## Quickstart
+
+### Installing and running Langtrain CLI
+
+Install globally with npm:
+
+```shell
+# Install using npm
+npm install -g langtrain
+```
+
+Then simply run `langtrain` to get started with the interactive menu.
+
+```shell
+langtrain
 ```
 
-## Advanced Usage
+<details>
+<summary><strong>Troubleshooting Installation</strong></summary>
+
+If you encounter permission errors, try running with sudo or fix your npm permissions:
+```bash
+sudo npm install -g langtrain
+```
+</details>
 
-For more complex workflows, you can leverage specific modules:
+### Using Langtrain with your Cloud Plan
 
-### Fine-Tuning a Model
+Run `langtrain login` and enter your API Key from the dashboard to authenticate. We recommend signing into your Langtrain account to use **Cloud Finetuning**, **Agent Persistence**, and **Model Hosting** as part of your Pro or Enterprise plan. [Learn more about Langtrain Plans](https://langtrain.ai/pricing).
 
-```typescript
-// Start a fine-tuning job
-const job = await tune.createFineTune({
-  trainingFile: 'file-id-123',
-  model: 'gpt-3.5-turbo',
-  hyperparameters: {
-    nEpochs: 4
-  }
-});
+You can check your current subscription status and limits directly from the CLI:
 
-console.log(`Fine-tuning started: ${job.id}`);
+```shell
+langtrain status
 ```
 
-## Documentation
+*Free plans allow local fine-tuning and limited cloud agent interactions. Upgrade to Pro for cloud-based GPU training.*
+
+## Features
+
+- **🤖 AI Agents**: Create, manage, and chat with custom AI agents hosted on Langtrain Server.
+- **🧠 Langtune**: Fine-tune LLMs (Llama 3, Mistral) locally or on the cloud.
+- **👁️ Langvision**: Optimize and fine-tune multimodal vision models.
+- **☁️ Data Persistence**: Automatically sync datasets and training jobs with your workspace.
 
-Visit [docs.langtrain.ai](https://docs.langtrain.ai) for comprehensive guides and API reference.
+## Docs
 
-## License
+- [**Langtrain Documentation**](https://docs.langtrain.ai)
+- [**SDK Reference**](https://docs.langtrain.ai/sdk)
+- [**Contributing**](./CONTRIBUTING.md)
 
-MIT © [Langtrain AI](https://langtrain.ai)
+This repository is licensed under the [MIT License](LICENSE).

package/dist/cli.d.mts

@@ -0,0 +1 @@
+#!/usr/bin/env node

package/dist/cli.d.ts

@@ -0,0 +1 @@
+#!/usr/bin/env node