openchat-7b

MAX Model

1 versions

A family of open-source models trained on a wide variety of data, surpassing ChatGPT on various benchmarks. Updated to version 3.5-0106.

Run this model

  1. Install our magic package manager:

    curl -ssL https://magic.modular.com/ | bash

    Then run the source command that's printed in your terminal.

  2. Install Max Pipelines in order to run this model.

    magic global install max-pipelines
  3. Start a local endpoint for openchat/7b:

    max-serve serve --huggingface-repo-id openchat/openchat_3.5

    The endpoint is ready when you see the URI printed in your terminal:

    Uvicorn running on http://0.0.0.0:8000 (Press CTRL+C to quit)
  4. Now open another terminal to send a request using curl:

    curl -N http://0.0.0.0:8000/v1/chat/completions -H "Content-Type: application/json" -d '{
        "model": "openchat/7b",
        "stream": true,
        "messages": [
            {"role": "system", "content": "You are a helpful assistant."},
            {"role": "user", "content": "Who won the World Series in 2020?"}
        ]
    }' | grep -o '"content":"[^"]*"' | sed 's/"content":"//g' | sed 's/"//g' | tr -d '
    ' | sed 's/\n/
    /g'
  5. 🎉 Hooray! You’re running Generative AI. Our goal is to make this as easy as possible.

About

OpenChat is a series of cutting-edge open-source language models fine-tuned using a novel approach called C-RLFT. This method is inspired by offline reinforcement learning, enabling the models to achieve advanced performance across a variety of tasks.

The latest iteration, OpenChat-3.5-1210, demonstrates exceptional proficiency in coding tasks and achieves high scores on numerous open-source LLM benchmarks. This makes it a powerful tool for developers and researchers seeking advanced AI capabilities in open-source ecosystems.

References

GitHub

HuggingFace

DETAILS

MODEL CLASS
MAX Model

MAX Models are extremely optimized inference pipelines to run SOTA performance for that model on both CPU and GPU. For many of these models, they are the fastest version of this model in the world.

Browse 18+ MAX Models

MODULAR GITHUB

Modular

CREATED BY

openchat

MODEL

openchat/openchat_3.5

TAGS

C-RLFT
arxiv:2303.08774
arxiv:2309.11235
autotrain_compatible
conversational
dataset:LDJnr/LessWrong-Amplify-Instruct
dataset:LDJnr/Pure-Dove
dataset:LDJnr/Verified-Camel
dataset:OpenAssistant/oasst_top1_2023-08-25
dataset:TIGER-Lab/MathInstruct
dataset:glaiveai/glaive-code-assistant
dataset:imone/OpenOrca_FLAN
dataset:meta-math/MetaMathQA
dataset:openchat/openchat_sharegpt4_dataset
dataset:tiedong/goat
endpoints_compatible
license:apache-2.0
mistral
openchat
pytorch
region:us
text-generation
text-generation-inference
transformers

@ Copyright - Modular Inc - 2024