ritabratamaiti/instructmix-llama-3b – Run with an API on Replicate

ritabratamaiti / instructmix-llama-3b

InstructMix Llama 3B is a language model fine-tuned on the InstructMix dataset.

Cold

Public
169 runs
L40S

Iterate in playground

Run with an API

Playground API Examples README Versions

Input

instruction

*string

Shift + Return to add a new line

Describes the task the model should perform. Must be provided if format_prompt is True.

input

string

Shift + Return to add a new line

Optional context or input for the task. Used to generate a prompt if format_prompt is True.

response_prefix

string

Shift + Return to add a new line

Optional response prefix. It will be added to the beginning of the response and will help guide the response generation.

temperature

number

(minimum: 0.01, maximum: 5)

Adjusts randomness of outputs, greater than 1 is random and 0 is deterministic. (minimum: 0.01; maximum: 5)

Default: 0.1

top_p

number

(minimum: 0.01, maximum: 1)

When decoding text, samples from the top p percentage of most likely tokens; lower to ignore less likely tokens (minimum: 0.01; maximum: 1)

Default: 0.75

top_k

integer

(minimum: 1, maximum: 100)

The number of highest probability vocabulary tokens to keep for top-k-filtering (minimum: 1; maximum: 100)

Default: 40

num_beams

integer

(minimum: 1, maximum: 10)

Number of beams for beam search (minimum: 1; maximum: 10)

Default: 4

repetition_penalty

number

(minimum: 0.01, maximum: 5)

Penalize repeated words. (minimum: 0.01; maximum: 5)

Default: 1.1

max_new_tokens

integer

(minimum: 1, maximum: 512)

Maximum number of new tokens to generate. A word is 2-3 tokens. (minimum: 1; maximum: 512)

Default: 128

Run this model in Node.js with one line of code:

npx create-replicate --model=ritabratamaiti/instructmix-llama-3b

or set up a project from scratch

Install Replicate’s Node.js client library:

npm install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run ritabratamaiti/instructmix-llama-3b using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "ritabratamaiti/instructmix-llama-3b:a5631f2884c9e16bc7018f060da3e63c1eef0e549e97ff21337f8605b6dab24e",
  {
    input: {
      top_k: 40,
      top_p: 0.75,
      num_beams: 4,
      instruction: "What is the meaning of life?",
      temperature: 0.1,
      max_new_tokens: 128,
      repetition_penalty: 1.1
    }
  }
);

console.log(output);

To learn more, take a look at the guide on getting started with Node.js.

Install Replicate’s Python client library:

pip install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import the client:

import replicate

Run ritabratamaiti/instructmix-llama-3b using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "ritabratamaiti/instructmix-llama-3b:a5631f2884c9e16bc7018f060da3e63c1eef0e549e97ff21337f8605b6dab24e",
    input={
        "top_k": 40,
        "top_p": 0.75,
        "num_beams": 4,
        "instruction": "What is the meaning of life?",
        "temperature": 0.1,
        "max_new_tokens": 128,
        "repetition_penalty": 1.1
    }
)
print(output)

To learn more, take a look at the guide on getting started with Python.

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Run ritabratamaiti/instructmix-llama-3b using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "ritabratamaiti/instructmix-llama-3b:a5631f2884c9e16bc7018f060da3e63c1eef0e549e97ff21337f8605b6dab24e",
    "input": {
      "top_k": 40,
      "top_p": 0.75,
      "num_beams": 4,
      "instruction": "What is the meaning of life?",
      "temperature": 0.1,
      "max_new_tokens": 128,
      "repetition_penalty": 1.1
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

Output

The meaning of life is an age-old question that has been debated by philosophers, scientists, and religious leaders for centuries. Some believe that life has a purpose, while others believe that it is simply a random occurrence. Ultimately, the meaning of life is a personal journey that each individual must discover for themselves.

Generated in

17.3 seconds

Tweak it Report

This output was created using a different version of the model, ritabratamaiti/instructmix-llama-3b:037316a9.

Run time and cost

This model costs approximately $0.0037 to run on Replicate, or 270 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on Nvidia L40S GPU hardware. Predictions typically complete within 4 seconds. The predict time for this model varies significantly based on the inputs.

Readme

InstructMix Llama 3B is a language model fine-tuned on the InstructMix dataset. It is an instruction-tuned, 3-billion-parameter model and follows the alpaca format.