Ollama Library Scraper

Programmatically access Ollama's model library with a type-safe TypeScript API. Search models, extract metadata, compare variants, and retrieve download statistics—no HTML parsing required.

Installation

npm install ollama-library-scraper
yarn add ollama-library-scraper
pnpm add ollama-library-scraper
bun add ollama-library-scraper

Quick Start

import { OllamaScraper } from 'ollama-library-scraper';

const scraper = new OllamaScraper();

// Get all models
const models = await scraper.getModelListing();
console.log(`Found ${models.length} models`);

// Get specific model details
const details = await scraper.getModelDetails('llama3.1');
console.log(details);

Usage

Basic Usage

// ES Modules
import { OllamaScraper } from 'ollama-library-scraper';

// CommonJS
const { OllamaScraper } = require('ollama-library-scraper');

const scraper = new OllamaScraper();

// Get all models
const models = await scraper.getModelListing();
console.log(`Found ${models.length} models`);

// Search for models
const qwenModels = await scraper.getModelListing({
  query: 'qwen',
  sort: 'newest',
});

// Get model details
const details = await scraper.getModelDetails('qwen3-coder');
console.log(details);

// Get model tags
const tags = await scraper.getModelTags('qwen3-coder');
console.log(`Model has ${tags.length} tags`);

Advanced Examples

// Search with different sort options
const popularModels = await scraper.getModelListing({
  sort: 'most-popular',
});

const newestCodeModels = await scraper.getModelListing({
  query: 'code',
  sort: 'newest',
});

// Get details for a specific tag
const tagDetails = await scraper.getModelDetails('qwen3-coder', '30b-a3b-q4_K_M');

// Filter models by capabilities
const visionModels = (await scraper.getModelListing()).filter(model =>
  model.capabilities?.includes('vision')
);

const toolsModels = (await scraper.getModelListing()).filter(model =>
  model.capabilities?.includes('tools')
);

Real-World Examples

Find the most popular code models

const popularCodeModels = await scraper.getModelListing({
  query: 'code',
  sort: 'most-popular',
});

console.log('Top code models:', popularCodeModels.slice(0, 5));

Get all vision-capable models

const allModels = await scraper.getModelListing();
const visionModels = allModels.filter(model => model.capabilities?.includes('vision'));

console.log(`Found ${visionModels.length} vision models`);

Compare model sizes across variants

const modelName = 'llama3.1';
const tags = await scraper.getModelTags(modelName);

tags.forEach(tag => {
  console.log(`${tag.name}: ${tag.size}`);
});

Build a model recommendation system

async function recommendModels(userQuery: string) {
  const models = await scraper.getModelListing({
    query: userQuery,
    sort: 'most-popular',
  });

  return models.slice(0, 10).map(model => ({
    name: model.name,
    description: model.description,
    popularity: model.pulls,
    capabilities: model.capabilities || [],
  }));
}

const recommendations = await recommendModels('chat');

API Reference

OllamaScraper Class

getModelListing(options?)

Fetches the list of models from the Ollama library.

Parameters:

• options (optional):
  • query?: string - Search query to filter models
  • sort?: SortOption - Sort order ('newest' | 'most-popular' | 'oldest' | 'alphabetical')

Returns: Promise<ModelListItem[]>

getModelDetails(modelName, tag?)

Gets detailed information about a specific model.

Parameters:

• modelName: string - The name of the model (e.g., 'qwen3-coder')
• tag?: string - Optional specific tag/version (e.g., '30b-a3b-q4_K_M')

Returns: Promise<ModelDetails>

getModelTags(modelName)

Retrieves all available tags for a specific model.

Parameters:

• modelName: string - The name of the model

Returns: Promise<ModelTag[]>

Data Types

ModelListItem

interface ModelListItem {
  name: string; // e.g., "qwen3-coder"
  description: string; // Model description
  parameters: string[]; // e.g., ["30b", "480b"]
  pulls: string; // e.g., "17.9K"
  tags: number; // Number of available tags
  lastUpdated: string; // e.g., "2 days ago"
  url: string; // Full URL to model page
  capabilities?: string[]; // e.g., ["tools", "vision", "thinking", "embedding"]
}

ModelDetails

interface ModelDetails {
  name: string;
  description: string;
  downloads: string; // e.g., "18K"
  lastUpdated: string;
  readmeHtml?: string; // Raw HTML content from readme section
  readmeMarkdown?: string; // Converted Markdown content
  models: {
    name: string; // Tag name
    size: string; // e.g., "19GB"
    contextWindow?: string; // e.g., "256K"
    inputType?: string; // e.g., "Text", "Vision"
    lastUpdated: string;
  }[];
}

ModelTag

interface ModelTag {
  name: string; // Tag name
  size: string; // Model size (e.g., "19GB")
  digest: string; // Model digest hash (12 chars)
  modifiedAt: string; // Last modified date
  contextWindow?: string; // e.g., "256K"
  inputType?: string; // e.g., "Text", "Vision"
  aliases?: string[]; // e.g., ["latest"]
}

SortOption

type SortOption = 'newest' | 'most-popular' | 'oldest' | 'alphabetical';

Development

Setup

# Clone the repository
git clone https://github.com/helgesverre/ollama-library-scraper.git
cd ollama-library-scraper

# Install dependencies
npm install

# Run development mode
npm run dev

Scripts

• npm run build - Build the TypeScript project
• npm run dev - Run the example script in development mode
• npm test - Run tests
• npm run test:watch - Run tests in watch mode
• npm run test:coverage - Run tests with coverage report
• npm run format - Format all code files with Prettier
• npm run format:check - Check code formatting
• npm run lint - Run linting and formatting checks

Testing

The project uses Vitest for testing. Tests mock the HTTP requests to avoid hitting the actual website during testing.

# Run tests
npm test

# Run tests with coverage
npm run test:coverage

Error Handling

All methods throw errors with descriptive messages:

try {
  const models = await scraper.getModelListing();
} catch (error) {
  console.error('Failed to fetch models:', error.message);
}

Why Use This Library?

• Avoid Manual HTML Parsing - Pre-built selectors and parsing logic
• Type Safety - Full TypeScript definitions prevent runtime errors
• Maintained - Regularly updated to match ollama.com structure changes
• Well Tested - Comprehensive test suite ensures reliability
• Simple API - Three main methods cover all use cases
• Framework Agnostic - Works with any Node.js framework or runtime
• No Official API - Ollama doesn't provide an API for their model library; this is the only programmatic way to access it

Important Notes

• ⚠️ No Rate Limiting: Implement your own delays to be respectful of ollama.com servers
• 🔄 HTML Changes: The scraper depends on ollama.com's HTML structure, which may change
• 📦 No Caching: Each call makes a fresh HTTP request - implement caching if needed
• 🌐 Network Required: All methods require internet connectivity
• 🛡️ Error Handling: Always wrap calls in try-catch blocks for production use

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add some amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

License

This project is licensed under the MIT License - see the LICENSE file for details.