Providers

llmist supports multiple LLM providers out of the box.

Supported Providers

Provider	Env Variable	Prefix	Reasoning	Pricing
OpenAI	OPENAI_API_KEY	openai:	✓ reasoning.effort	Paid
Anthropic	ANTHROPIC_API_KEY	anthropic:	✓ Extended thinking	Paid
Google Gemini	GEMINI_API_KEY	gemini:	✓ Thinking config	Paid
OpenRouter	OPENROUTER_API_KEY	openrouter: or or:	✓ (model-dependent)	Paid
HuggingFace	HF_TOKEN	huggingface: or hf:	—	Free

Auto-Discovery

llmist automatically discovers providers based on environment variables:

export OPENAI_API_KEY="sk-..."
export ANTHROPIC_API_KEY="sk-ant-..."
export GEMINI_API_KEY="..."
export HF_TOKEN="hf_..."

const client = new LLMist();

.withModel('gpt-5')                                    // OpenAI (auto-detected)
.withModel('claude-sonnet-4-5')                        // Anthropic (auto-detected)
.withModel('gemini-2.5-flash')                         // Gemini (auto-detected)
.withModel('meta-llama/Llama-3.1-8B-Instruct')         // HuggingFace (auto-detected)

Explicit Provider Prefix

Use provider:model format:

.withModel('openai:gpt-5')
.withModel('anthropic:claude-sonnet-4-5-20250929')
.withModel('gemini:gemini-2.5-flash')
.withModel('huggingface:deepseek-ai/DeepSeek-V3.2')
.withModel('hf:Qwen/Qwen2.5-72B-Instruct:fastest')  // With routing

Manual Provider Setup

import { LLMist, OpenAIChatProvider, AnthropicMessagesProvider, OpenRouterProvider } from 'llmist';
import OpenAI from 'openai';

const client = new LLMist({
  autoDiscoverProviders: false,
  adapters: [
    new OpenAIChatProvider({ apiKey: 'sk-...' }),
    new AnthropicMessagesProvider({ apiKey: 'sk-ant-...' }),
  ],
  defaultProvider: 'openai',
});

OpenRouterProvider

OpenRouterProvider provides access to 400+ models from dozens of providers through a single unified gateway. It supports prompt caching, model routing strategies, and reasoning models.

import { LLMist, OpenRouterProvider } from 'llmist';
import OpenAI from 'openai';

const openrouterClient = new OpenAI({
  apiKey: 'sk-or-...',
  baseURL: 'https://openrouter.ai/api/v1',
  defaultHeaders: {
    'HTTP-Referer': 'https://myapp.com',  // Optional: for analytics
    'X-Title': 'My App',                  // Optional: for analytics
  },
});

const client = new LLMist({
  autoDiscoverProviders: false,
  adapters: [
    new OpenRouterProvider(openrouterClient, {
      siteUrl: 'https://myapp.com',   // Optional
      appName: 'My App',              // Optional
    }),
  ],
});

Use the openrouter: or or: prefix to route to specific models, with optional routing strategy:

.withModel('openrouter:anthropic/claude-sonnet-4-5')
.withModel('or:meta-llama/llama-3.1-70b-instruct:fastest')  // Route to fastest provider
.withModel('or:mistralai/mistral-large:cheapest')             // Route to cheapest provider

Creating Custom Providers

interface ProviderAdapter {
  readonly providerId: string;
  readonly priority?: number;

  supports(model: ModelDescriptor): boolean;
  stream(options: LLMGenerationOptions, descriptor: ModelDescriptor): LLMStream;
  getModelSpecs?(): ModelSpec[];
  countTokens?(messages: LLMMessage[], descriptor: ModelDescriptor): Promise<number>;
}

See Also

• Model Catalog - Query available models
• Custom Models - Register fine-tuned models
• Reasoning Models - Provider-specific reasoning support