@anolilab/ai-model-registry
Advanced tools
Unified registry for AI model providers and their metadata
A comprehensive, unified registry for AI model providers and their metadata. This package provides a tree-shakable interface to access model information from 50+ AI providers including OpenAI, Anthropic, Google, Meta, Groq, and many others.
# Using npm
npm install @anolilab/ai-model-registry
# Using yarn
yarn add @anolilab/ai-model-registry
# Using pnpm
pnpm add @anolilab/ai-model-registry
import { getProviders, getModelsByProvider, getModelById, searchModels, getAllModels } from "@anolilab/ai-model-registry";
// Get all available providers
const providers = getProviders();
console.log(providers);
// ['Anthropic', 'Google', 'Groq', 'Meta', 'OpenAI', 'DeepSeek', ...]
// Get all models from a specific provider
const anthropicModels = getModelsByProvider("Anthropic");
console.log(`Found ${anthropicModels.length} Anthropic models`);
// Get a specific model by ID
const model = getModelById("claude-3-opus-20240229");
if (model) {
console.log(`Model: ${model.name}`);
console.log(`Provider: ${model.provider}`);
console.log(`Cost: $${model.cost.input}/1K input, $${model.cost.output}/1K output`);
console.log(`Context: ${model.limit.context?.toLocaleString()} tokens`);
}
// Search for models with specific capabilities
const visionModels = searchModels({ vision: true });
const reasoningModels = searchModels({ reasoning: true });
const toolCallModels = searchModels({ tool_call: true });
// Get all models for advanced filtering
const allModels = getAllModels();
getProviders(): string[]Returns an array of all available provider names.
const providers = getProviders();
// ['Anthropic', 'Google', 'Groq', 'Meta', 'OpenAI', ...]
getModelsByProvider(provider: string): Model[]Returns all models for a specific provider.
const openAIModels = getModelsByProvider("OpenAI");
const anthropicModels = getModelsByProvider("Anthropic");
getModelById(id: string): Model | undefinedReturns a specific model by its ID, or undefined if not found.
const gpt4 = getModelById("gpt-4");
const claude = getModelById("claude-3-opus-20240229");
getAllModels(): Model[]Returns all models (useful for advanced filtering and custom logic).
const allModels = getAllModels();
const expensiveModels = allModels.filter((model) => (model.cost.input || 0) > 0.1 || (model.cost.output || 0) > 0.1);
getProviderStats(): Record<string, number>Returns provider statistics with model counts.
const stats = getProviderStats();
console.log(stats);
// {
// 'OpenAI': 15,
// 'Anthropic': 8,
// 'Google': 12,
// 'Meta': 25,
// ...
// }
searchModels(criteria: SearchCriteria): Model[]Search models by various criteria with powerful filtering options.
interface SearchCriteria {
// Capability filters
vision?: boolean;
reasoning?: boolean;
tool_call?: boolean;
streaming_supported?: boolean;
preview?: boolean;
// Provider filter
provider?: string;
// Modality filters
modalities?: {
input?: string[];
output?: string[];
};
// Context window filters
context_min?: number;
context_max?: number;
// Cost filters
max_input_cost?: number;
max_output_cost?: number;
}
// Find all vision-capable models
const visionModels = searchModels({ vision: true });
// Find models with reasoning capabilities
const reasoningModels = searchModels({ reasoning: true });
// Find models that support tool calling
const toolCallModels = searchModels({ tool_call: true });
// Find models from a specific provider
const openAIModels = searchModels({ provider: "OpenAI" });
// Find models with large context windows
const largeContextModels = searchModels({ context_min: 100000 });
// Find affordable models
const affordableModels = searchModels({
max_input_cost: 0.01,
max_output_cost: 0.02,
});
// Find models that accept text and image input
const multimodalModels = searchModels({
modalities: {
input: ["text", "image"],
},
});
// Find models with streaming support
const streamingModels = searchModels({ streaming_supported: true });
// Find preview/beta models
const previewModels = searchModels({ preview: true });
Each model follows a comprehensive schema with the following structure:
interface Model {
// Core identification
id: string;
name: string | null;
provider?: string;
providerId?: string;
// Provider metadata
providerEnv?: string[];
providerNpm?: string;
providerDoc?: string;
providerModelsDevId?: string;
// Date information
releaseDate?: string | null;
lastUpdated?: string | null;
launchDate?: string;
trainingCutoff?: string | null;
// Capabilities
attachment: boolean;
reasoning: boolean;
temperature: boolean;
toolCall: boolean;
openWeights: boolean;
vision?: boolean;
extendedThinking?: boolean;
preview?: boolean;
// Knowledge and context
knowledge?: string | null;
// Pricing structure
cost: {
input: number | null; // per 1K tokens
output: number | null; // per 1K tokens
inputCacheHit: number | null; // cache hit pricing
imageGeneration?: number | null;
imageGenerationUltra?: number | null;
videoGeneration?: number | null;
videoGenerationWithAudio?: number | null;
videoGenerationWithoutAudio?: number | null;
};
// Limits
limit: {
context: number | null; // max tokens
output: number | null; // max tokens
};
// Modalities
modalities: {
input: string[]; // ['text', 'image', 'audio', ...]
output: string[]; // ['text', 'image', 'audio', ...]
};
// Infrastructure
regions?: string[];
streamingSupported?: boolean | null;
deploymentType?: string;
version?: string | null;
// Provider-specific capabilities
cacheRead?: boolean;
codeExecution?: boolean;
searchGrounding?: boolean;
structuredOutputs?: boolean;
batchMode?: boolean;
audioGeneration?: boolean;
imageGeneration?: boolean;
compoundSystem?: boolean;
// Version management
versions?: {
stable?: string | null;
preview?: string | null;
};
// Additional metadata
description?: string;
ownedBy?: string;
originalModelId?: string;
providerStatus?: string;
supportsTools?: boolean;
supportsStructuredOutput?: boolean;
}
The package supports tree shaking, so you can import only what you need:
// Only import specific functions
import { getProviders, getModelById } from "@anolilab/ai-model-registry";
// Import schema for validation
import { ModelSchema } from "@anolilab/ai-model-registry/schema";
// Import icons (if needed)
import { getIcon } from "@anolilab/ai-model-registry/icons";
The registry includes models from 50+ providers:
And many more...
# Clone the repository
git clone https://github.com/anolilab/ai-models.git
cd ai-models
# Install dependencies
pnpm install
# Complete build process
pnpm run download # Download provider data
pnpm run aggregate # Aggregate and enrich data
pnpm run generate-icons # Generate provider icons
pnpm run build # Build the package
# Run tests
pnpm test
The complete build process involves several steps to download, aggregate, and generate all necessary data:
# 1. Download provider data
pnpm run download
# 2. Aggregate and enrich provider data (includes pricing from Helicone API)
pnpm run aggregate
# 3. Generate provider icons
pnpm run generate-icons
# 4. Build the package
pnpm run build
Download (pnpm run download): Downloads model data from various AI providers using transformers in scripts/download/transformers/. This creates the raw provider data in data/providers/.
Aggregate (pnpm run aggregate):
data/providers/data/all-models.json and src/models-data.tsGenerate Icons (pnpm run generate-icons):
src/icons-sprite.ts with icon mappingsBuild (pnpm run build):
# Download provider data from various sources
pnpm run download
# Download data for a specific provider
pnpm run download --provider openai
pnpm run download --provider anthropic
# Aggregate provider data (includes pricing enrichment and synchronization)
pnpm run aggregate
# Generate provider icons
pnpm run generate-icons
# Build the package
pnpm run build
# Build for production
pnpm run build:prod
# Run tests
pnpm test
# Run tests with coverage
pnpm run test:coverage
# Lint code
pnpm run lint:eslint
# Type check
pnpm run lint:types
packages/ai-model-registry/
├── src/
│ ├── index.ts # Main exports
│ ├── schema.ts # Model schema definitions
│ └── models-data.ts # Generated model data
├── scripts/
│ ├── aggregate-providers.ts # Data aggregation script
│ ├── generate-svg-sprite.ts # Icon generation
│ └── download/ # Provider data downloaders
├── data/
│ ├── all-models.json # Generated model data
│ └── providers/ # Raw provider data
└── assets/
└── icons/ # Provider icons
This package automatically includes real-time pricing data from Helicone's LLM Cost API during the aggregation process.
Helicone provides pricing data for models from:
# Aggregate all models with pricing data
pnpm run aggregate
# Build the package (includes aggregation with pricing)
pnpm run build
The provider registry includes a powerful data synchronization system that automatically merges missing data between models with the same ID across different providers.
The following cost-related fields are never synchronized to preserve pricing accuracy:
cost (entire cost object)input (input cost)output (output cost)inputCacheHit (cache hit pricing)imageGeneration (image generation pricing)videoGeneration (video generation pricing)If you have the same model (e.g., gpt-4) from multiple providers:
OpenAI Provider:
{
"id": "gpt-4",
"name": "GPT-4",
"cost": { "input": 0.03, "output": 0.06 },
"description": null,
"releaseDate": "2023-03-14"
}
Azure Provider:
{
"id": "gpt-4",
"name": null,
"cost": { "input": 0.03, "output": 0.06 },
"description": "GPT-4 is a large multimodal model",
"releaseDate": null
}
Result after synchronization:
{
"id": "gpt-4",
"name": "GPT-4",
"cost": { "input": 0.03, "output": 0.06 },
"description": "GPT-4 is a large multimodal model",
"releaseDate": "2023-03-14"
}
We welcome contributions! Please see our Contributing Guide for details.
scripts/download/transformers/scripts/config.tspnpm run download --provider <provider-name>pnpm run aggregatePlease use our Issue Tracker to report bugs or request features.
This project is licensed under the Apache License 2.0 - see the LICENSE file for details.
Made with ❤️ by AnoliLab
FAQs
Unified registry for AI model providers and their metadata
We found that @anolilab/ai-model-registry demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.
Did you know?
Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.
Research
/Security News
Socket identified 80 fake candidates targeting engineering roles, including suspected North Korean operators, exposing the new reality of hiring as a security function.
Application Security
/Research
/Security News
Socket detected multiple compromised CrowdStrike npm packages, continuing the "Shai-Hulud" supply chain attack that has now impacted nearly 500 packages.
Research
/Security News
Malicious update to @ctrl/tinycolor on npm is part of a supply-chain attack hitting 40+ packages across maintainers