Text-Based Models

To utilize text-based models effectively in your application using the Merlin API, follow this guide to understand the available models and how to interact with them using Node.js.

Available Models

Below is the list of the available text-based model IDs you can use with the Merlin API along with a brief description:

Model ID	Description	Provider	Pricing (in $ per 1K tokens)
gpt-3.5-turbo	A cost-effective and fast model, suitable for a wide range of tasks.	OpenAI	Pricing
gpt-4-1106-preview	An early preview version of the GPT-4 model.	OpenAI	Pricing
gpt-4	The latest generation GPT-4 model, offering advanced capabilities.	OpenAI	Pricing
claude-instant-1	A model from Anthropic, designed for instant completions.	Anthropic	Pricing
claude-2	The second version of the Claude model from Anthropic.	Anthropic	Pricing
codellama-34b	A fine-tuned model specialized in generating code-related content.	Meta	0.001
mistral-7b	A medium-sized, instruction-following version of MistralAI.	Mistral	0.00015
mistral-7b-openorca	Fine tuned mistral model trained by Orca Foundation	Orca Foundation	0.00015
zephyr-7b-beta	A prototype version of a large-scale model from HuggingFace.	Huggingface	0.00015
llama-2-7b	Small sized open source model by Meta	Meta	0.00015
llama-2-13b	Medium sized open source model by Meta	Meta	0.00025
llama-2-70b	Largest open source model by Meta	Meta	0.001
gpt-4-0314	A version of GPT-4 with specialized optimizations.	OpenAI	Pricing
gpt-4-0613	A GPT-4 variant optimized for certain tasks.	OpenAI	Pricing
gpt-4-32k	A GPT-4 model with an extended token limit.	OpenAI	Pricing
gpt-4-32k-0314	A specialized version of GPT-4 with a 32k token limit.	OpenAI	Pricing
gpt-4-32k-0613	Another variant of GPT-4 with the 32k token limit, with different optimizations.	OpenAI	Pricing
gpt-3.5-turbo-1106	A variant of GPT-3.5-turbo for specific use-cases.	OpenAI	Pricing
gpt-3.5-turbo-16k	An expansion of GPT-3.5-turbo with a larger token limit.	OpenAI	Pricing
gpt-3.5-turbo-0301	A specific configuration of GPT-3.5-turbo.	OpenAI	Pricing
gpt-3.5-turbo-0613	Another GPT-3.5-turbo configuration for particular applications.	OpenAI	Pricing
gpt-3.5-turbo-16k-0613	Combines the 16k expansion with the 0613 optimizations.	OpenAI	Pricing
gemini-pro	Google's competitor to the GPT-4 model, with multimodal capabilities	Google

Interacting with Text-Based Models

To interact with the provided models using Node.js, you can use the following sample code:

import { Merlin } from "merlin-node";

const apiKey = "<YOUR_MERLIN_API_KEY>"; // Replace with your API key from Merlin
const merlin = new Merlin({ merlinConfig: { apiKey } });

async function createCompletion() {
  try {
    const completion = await merlin.chat.completions.create({
      messages: [{ role: "system", content: "You are a helpful assistant." }],
      model: "gpt-3.5-turbo", // Adjust model as needed
    });

    console.log(completion.choices[0].message.content);
  } catch (error) {
    console.error("Error creating completion:", error);
  }
}

createCompletion();

Remember to install the merlin-node package and add your API key to use Merlin API.

Notes

• Make sure to replace the model property's value with the desired model ID from the table above depending on your task requirements.
• The output from the createCompletion function call is logged and will visible in the console soon.