@llmgateway/models

Model and provider definitions for LLM Gateway - the unified API for all LLM providers.

Installation

npm install @llmgateway/models

Usage

Models

The models array contains all supported LLM models with their pricing and provider mappings:

import { models } from "@llmgateway/models";

// Find a specific model
const gpt4 = models.find((m) => m.id === "gpt-4o");

console.log(gpt4.id); // "gpt-4o"
console.log(gpt4.family); // "openai"
console.log(gpt4.contextSize); // 128000
console.log(gpt4.providers); // Provider mappings with pricing

Model Properties

| Property | Type | Description | | ------------------- | ------------------------ | ---------------------------------------------- | | id | string | Unique model identifier | | family | string | Model family (openai, anthropic, google, etc.) | | contextSize | number | Maximum context window in tokens | | maxOutput | number | Maximum output tokens | | providers | ProviderModelMapping[] | Available providers with pricing | | supportsImages | boolean | Whether model supports image inputs | | supportsTools | boolean | Whether model supports tool/function calling | | supportsReasoning | boolean | Whether model supports extended reasoning | | free | boolean | Whether model is free to use |

Provider Mapping Properties

| Property | Type | Description | | ------------------ | -------- | ---------------------------------- | | providerId | string | Provider identifier | | modelName | string | Model name at the provider | | inputPrice | number | Price per input token (USD) | | outputPrice | number | Price per output token (USD) | | cachedInputPrice | number | Price per cached input token (USD) | | contextSize | number | Provider-specific context size | | maxOutput | number | Provider-specific max output |

Providers

The providers array contains all supported LLM providers:

import { providers } from "@llmgateway/models";

// List all providers
providers.forEach((p) => {
  console.log(`${p.name}: ${p.description}`);
});

// Find a specific provider
const openai = providers.find((p) => p.id === "openai");
console.log(openai.name); // "OpenAI"
console.log(openai.website); // "https://openai.com"

Provider Properties

| Property | Type | Description | | ------------- | --------- | ----------------------------------- | | id | string | Unique provider identifier | | name | string | Display name | | description | string | Provider description | | website | string | Provider website URL | | streaming | boolean | Whether provider supports streaming | | color | string | Brand color (hex) |

Supported Providers

• OpenAI
• Anthropic
• Google AI Studio
• Google Vertex AI
• Azure OpenAI
• AWS Bedrock
• Mistral
• DeepSeek
• xAI (Grok)
• Perplexity
• And more...

Full Model List

For the complete list of models with live pricing, visit llmgateway.io/models.

License

See LICENSE for details.