# Hugging Face Guide

Welcome to the guide on Hugging Face, specifically focusing on the Orion-14B series models. In this guide, you will find detailed information about the Orion-14B series models, their features, benchmarks, and how to use them for various tasks.

## Introduction to Orion-14B Series Models

The Orion-14B series models are open-source multilingual large language models trained from scratch by OrionStarAI. These models have been trained on a 2.5T multilingual corpus, including languages such as Chinese, English, Japanese, Korean, and more. The base model of the Orion-14B series is trained on a diverse dataset of 2.5 trillion tokens, exhibiting superior performance in various languages.

### Features of Orion-14B Series Models

– Exceptional performance at a 20B-parameter scale level
– Strong multilingual capabilities, particularly excelling in Japanese and Korean
– Fine-tuned models demonstrate strong adaptability in human-annotated blind tests
– Long-chat version supports extremely lengthy texts
– Quantized versions reduce model size by 70% and improve inference speed by 30% with minimal performance loss

## Models in the Orion-14B Series

The Orion-14B series includes various models catering to different use-cases. Some of these models include:

1. Orion-14B-Base
2. Orion-14B-Chat
3. Orion-14B-LongChat
4. Orion-14B-Chat-RAG
5. Orion-14B-Chat-Plugin
6. Orion-14B-Base-Int4
7. Orion-14B-Chat-Int4

Detailed benchmarks and evaluations for these models are available in the guide.

## Model Release and Download

The models are available for download and use. You can find download links in the document.

## How to Use the Models

The guide provides detailed information on how to use the models for various tasks. This includes Python code, command-line tools, direct script inference, and example outputs for casual chat, Japanese & Korean chat, and more.

## Declarations and License

The guide includes important declarations and information about the license for the community use of the Orion-14B series models.

## Contact Information

For companies with a demand for deploying large-scale model applications, contact information for OrionStarAI is provided.

Thank you for reading this guide on Hugging Face and the Orion-14B series models. We hope this information helps you understand and utilize these powerful language models efficiently. If you have any further questions or require assistance, feel free to contact us.

Source link
The Orion-14B series models by OrionStarAI are open-source multilingual large language models that have been trained from scratch. These models have been trained on a diverse dataset of 2.5 trillion tokens, including languages such as Chinese, English, Japanese, Korean, and more. The following is a manual/tutorial for the Orion-14B series models.

1. Features of Orion-14B series models:
– Outstanding performance in comprehensive evaluations
– Strong multilingual capabilities, excelling in Japanese and Korean
– Strong adaptability in human-annotated blind tests
– Long-chat version supports extremely long texts, with exceptional performance
– Quantized versions reduce model size by 70% and improve inference speed by 30% with minimal performance loss

2. Orion-14B series models include:
– Orion-14B-Base
– Orion-14B-Chat
– Orion-14B-LongChat
– Orion-14B-Chat-RAG
– Orion-14B-Chat-Plugin
– Orion-14B-Base-Int4
– Orion-14B-Chat-Int4

3. Model release and download links are provided in the document.

4. Python Code and Command Line Tool are available for users to interact with the models.

5. Declarations and License are included to ensure responsible use of the models.

We hope this manual provides users with an understanding of the features and capabilities of the Orion-14B series models. For any inquiries or deployment of the models, companies are welcome to contact OrionStarAI.

  • Orion-14B series models are open-source multilingual large language models trained from scratch by OrionStarAI. The base model is trained on 2.5T multilingual corpus, including Chinese, English, Japanese, Korean, etc, and it exhibits superior performance in these languages. For details, please refer to tech report.

  • The Orion-14B series models exhibit the following features:

    • Among models with 20B-parameter scale level, Orion-14B-Base model shows outstanding performance in comprehensive evaluations.
    • Strong multilingual capabilities, significantly outperforming in Japanese and Korean testsets.
    • The fine-tuned models demonstrate strong adaptability, excelling in human-annotated blind tests.
    • The long-chat version supports extremely long texts, performing exceptionally well at a token length of 200k and can support up to a maximum of 320k.
    • The quantized versions reduce model size by 70%, improve inference speed by 30%, with performance loss less than 1%.
      opencompass modelcap
  • Orion-14B series models including:

    • Orion-14B-Base: A multilingual large language foundational model with 14 billion parameters, pretrained on a diverse dataset of 2.5 trillion tokens.
    • Orion-14B-Chat: A chat-model fine-tuned on a high-quality corpus aims to provide an excellence interactive experience for users in the large model community.
    • Orion-14B-LongChat: The long-context version excels at handling extremely lengthy texts, performing exceptionally well at a token length of 200k and can support up to a maximum of 320k.
    • Orion-14B-Chat-RAG: A chat-model fine-tuned on a custom retrieval augmented generation dataset, achieving superior performance in retrieval augmented generation tasks.
    • Orion-14B-Chat-Plugin: A chat-model specifically tailored for plugin and function calling tasks, ideal for agent-related scenarios where the LLM acts as a plugin and function call system.
    • Orion-14B-Base-Int4: A quantized base model utilizing 4-bit integer weights. It significantly reduces the model size by 70% and increases the inference speed by 30% while incurring a minimal performance loss of only 1%.
    • Orion-14B-Chat-Int4: A quantized chat model utilizing 4-bit integer weights.

Model release and download links are provided in the table below:

3.1. Base Model Orion-14B-Base Benchmarks

3.1.1. LLM evaluation results on examination and professional knowledge

Model C-Eval CMMLU MMLU AGIEval Gaokao BBH
LLaMA2-13B 41.4 38.4 55.0 30.9 18.2 45.6
Skywork-13B 59.1 61.4 62.7 43.6 56.1 48.3
Baichuan2-13B 59.0 61.3 59.5 37.4 45.6 49.0
QWEN-14B 71.7 70.2 67.9 51.9 62.5 53.7
InternLM-20B 58.8 59.0 62.1 44.6 45.5 52.5
Orion-14B-Base 72.9 70.6 69.9 54.7 62.1 56.5

3.1.2. LLM evaluation results on language understanding and common knowledge

Model RACE-middle RACE-high HellaSwag PIQA Lambada WSC
LLaMA 2-13B 63.0 58.9 77.5 79.8 76.5 66.3
Skywork-13B 87.6 84.1 73.7 78.3 71.8 66.3
Baichuan 2-13B 68.9 67.2 70.8 78.1 74.1 66.3
QWEN-14B 93.0 90.3 80.2 79.8 71.4 66.3
InternLM-20B 86.4 83.3 78.1 80.3 71.8 68.3
Orion-14B-Base 93.2 91.3 78.5 79.5 78.8 70.2

3.1.3. LLM evaluation results of OpenCompass testsets

Model Average Examination Language Knowledge Understanding Reasoning
LLaMA 2-13B 47.3 45.2 47.0 58.3 50.9 43.6
Skywork-13B 53.6 61.1 51.3 52.7 64.5 45.2
Baichuan 2-13B 49.4 51.8 47.5 48.9 58.1 44.2
QWEN-14B 62.4 71.3 52.67 56.1 68.8 60.1
InternLM-20B 59.4 62.5 55.0 60.1 67.3 54.9
Orion-14B-Base 64.3 71.4 55.0 60.0 71.9 61.6

3.1.4. Comparison of LLM performances on Japanese testsets

PLaMo-13B 52.3 56.7 42.8 95.8 70.6 71.0 8.70 70.5 2.40
WebLab-10B 50.7 66.6 53.7 82.1 62.9 56.2 10.0 72.0 2.40
ELYZA-jp-7B 48.8 71.7 25.3 86.6 70.8 64.1 2.50 62.1 7.20
StableLM-jp-7B 51.1 33.4 43.3 96.7 70.6 78.1 10.7 72.8 2.80
LLaMA 2-13B 46.3 75.0 47.6 38.8 76.1 67.7 18.1 63.2 10.4
Baichuan 2-13B 57.1 73.7 31.3 91.6 80.5 63.3 18.6 72.2 25.2
QWEN-14B 65.8 85.9 60.7 97.0 83.3 71.8 18.8 70.6 38.0
Yi-34B 67.1 83.8 61.2 95.2 86.1 78.5 27.2 69.2 35.2
Orion-14B-Base 69.1 88.2 75.8 94.1 75.7 85.1 17.3 78.8 38.0

3.1.5. Comparison of LLM performances on Korean testsets. n = 0 and n = 5 stand for n-shot prompts used in the evaluation

Model Average
n=0  n=5
n=0  n=5
n=0  n=5
n=0  n=5
n=0  n=5
KoGPT 53.0    70.1 55.9    58.3 73.5    72.9 45.1    59.8 37.5    89.4
Polyglot-ko-13B 69.6    73.7 59.5    63.1 79.4    81.1 48.2    60.4 91.2    90.2
LLaMA 2-13B 46.7    63.7 41.3    44.0 59.3    63.8 34.9    73.8 51.5    73.4
Baichuan 2-13B 52.1    58.7 39.2    39.6 60.6    60.6 58.4    61.5 50.3    72.9
QWEN-14B 53.8    73.7 45.3    46.8 64.9    68.9 33.4    83.5 71.5    95.7
Yi-34B 54.2    72.1 44.6    44.7 58.0    60.6 65.9    90.2 48.3    92.9
Orion-14B-Chat 74.5    79.6 47.0    49.6 77.7    79.4 81.6    90.7 92.4    98.7

3.1.6. Multilingual evaluation

Model Train Lang Japanese Korean Chinese English
PLaMo-13B En,Jp 52.3 * * *
Weblab-10B En,Jp 50.7 * * *
ELYZA-jp-7B En,Jp 48.8 * * *
StableLM-jp-7B En,Jp 51.1 * * *
KoGPT-6B En,Ko * 70.1 * *
Polyglot-ko-13B En,Ko * 70.7 * *
Baichuan2-13B Multi 57.1 58.7 50.8 57.1
Qwen-14B Multi 65.8 73.7 64.5 65.4
Llama2-13B Multi 46.3 63.7 41.4 55.3
Yi-34B Multi 67.1 72.2 58.7 68.8
Orion-14B-Chat Multi 69.1 79.5 67.9 67.3

3.2. Chat Model Orion-14B-Chat Benchmarks

3.2.1. Chat model subjective evaluation of MTBench

Model First-Turn Second-Turn Average
Baichuan2-13B-Chat 7.05 6.47 6.76
Qwen-14B-Chat 7.30 6.62 6.96
Llama2-13B-Chat 7.10 6.20 6.65
InternLM-20B-Chat 7.03 5.93 6.48
Orion-14B-Chat 7.68 7.07 7.37
* use vllm for inference

3.2.2. Chat model subjective evaluation of AlignBench

Model Math. Logi. Basic. Chi. Comp. Writ. Role. Prof. Avg.
Baichuan2-13B-Chat 3.76 4.07 6.22 6.05 7.11 6.97 6.75 6.43 5.25
Qwen-14B-Chat 4.91 4.71 6.90 6.36 6.74 6.64 6.59 6.56 5.72
Llama2-13B-Chat 3.05 3.79 5.43 4.40 6.76 6.63 6.99 5.65 4.70
InternLM-20B-Chat 3.39 3.92 5.96 5.50 7.18 6.19 6.49 6.22 4.96
Orion-14B-Chat 4.00 4.24 6.18 6.57 7.16 7.36 7.16 6.99 5.51
* use vllm for inference

3.3. LongChat Model Orion-14B-LongChat Benchmarks

3.3.1. LongChat evaluation of LongBench

Model NarrativeQA MultiFieldQA-en MultiFieldQA-zh DuReader QMSum VCSUM TREC TriviaQA LSHT RepoBench-P
GPT-3.5-Turbo-16k 23.60 52.30 61.20 28.70 23.40 16.00 68.00 91.40 29.20 53.60
LongChat-v1.5-7B-32k 16.90 41.40 29.10 19.50 22.70 9.90 63.50 82.30 23.20 55.30
Vicuna-v1.5-7B-16k 19.40 38.50 43.00 19.30 22.80 15.10 71.50 86.20 28.80 43.50
Yi-6B-200K 14.11 36.74 22.68 14.01 20.44 8.08 72.00 86.61 38.00 63.29
Orion-14B-LongChat 19.47 48.11 55.84 37.02 24.87 15.44 77.00 89.12 45.50 54.31

3.4. Chat RAG Model Benchmarks

3.4.1. LLM evaluation results of self-built RAG testsets

Model Effectiveness of Response(Keyword) *Effectiveness of Response(subjective evaluation) Quoting Ability Fallback Ability *AutoQA *Data Extraction
Baichuan2-13B-Chat 85 76 1 0 69 51
Qwen-14B-Chat 79 77 75 47 68 72
Qwen-72B-Chat(Int4) 87 89 90 32 67 76
GPT-4 91 94 96 95 75 86
Orion-14B-Chat-RAG 86 87 91 97 73 71
* means manual assessment

3.5. Chat Plugin Model Orion-14B-Chat-Plugin Benchmarks

3.5.1. LLM evaluation results of self-built plugin testsets

Model Intent Recognition with Full Params Intent Recognition with Missing Params Non-Plugin Invocation Recognition
Baichuan2-13B-Chat 25 0 0
Qwen-14B-Chat 55 0 50
GPT-4 95 52.38 70
Orion-14B-Chat-Plugin 92.5 60.32 90

3.6. Quantized Model Orion-14B-Base-Int4 Benchmarks

3.6.1. Comparison of before and after quantization

Model Size(GB) Inference Speed(tokens/s) C-Eval CMMLU MMLU RACE HellaSwag
OrionStar-14B-Base 28.0 135 72.8 70.6 70.0 93.3 78.5
OrionStar-14B-Base-Int4 8.3 178 71.8 69.8 69.2 93.1 78.0

Model weights, source code, and configuration needed for inference are published on Hugging Face, and the download link
is available in the table at the beginning of this document. We demonstrate various inference methods here, and the
program will automatically download the necessary resources from Hugging Face.

4.1. Python Code

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
from transformers.generation.utils import GenerationConfig

tokenizer = AutoTokenizer.from_pretrained("OrionStarAI/Orion-14B", use_fast=False, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("OrionStarAI/Orion-14B", device_map="auto",
                                             torch_dtype=torch.bfloat16, trust_remote_code=True)

model.generation_config = GenerationConfig.from_pretrained("OrionStarAI/Orion-14B")
messages = [{"role": "user", "content": "Hello, what is your name? "}]
response = model.chat(tokenizer, messages, streaming=False)

In the above Python code, the model is loaded with device_map='auto' to utilize all available GPUs. To specify the
device, you can use something like export CUDA_VISIBLE_DEVICES=0,1 (using GPUs 0 and 1).

4.2. Command Line Tool

CUDA_VISIBLE_DEVICES=0 python cli_demo.py

This command-line tool is designed for chat scenarios, and thus, it does not support calling the base model.

4.3. Direct Script Inference

# base model
CUDA_VISIBLE_DEVICES=0 python demo/text_generation_base.py --model OrionStarAI/Orion-14B --tokenizer OrionStarAI/Orion-14B --prompt hello

# chat model
CUDA_VISIBLE_DEVICES=0 python demo/text_generation.py --model OrionStarAI/Orion-14B-Chat --tokenizer OrionStarAI/Orion-14B-Chat --prompt hi

4.4 Example Output

4.4.1. Casual Chat

User: Hello
Orion-14B: Hello! How can I help you today? If you have any questions or need assistance, feel free to ask.
User: What functions do you have?
Orion-14B: As an AI language model, I have a wide range of functions and capabilities, including but not limited to:
  Answering general knowledge questions: I can provide information on various topics such as history, science, geography, and more.
  Providing definitions and explanations: I can explain complex concepts and provide definitions for words and phrases.
  Assisting with tasks: I can help with simple tasks like setting reminders, converting units of measurement, and even ordering food from restaurants.
  Generating text: I can write short stories, essays, articles, and other types of content based on prompts given to me.
  Translating languages: I can translate text from one language to another.
  Chatting: I can engage in conversation with users and respond to their queries in a conversational manner.
  Summarizing long texts: I can summarize longer texts like articles, books, and research papers.
  Generating random numbers and names: I can generate random numbers within a specified range and also suggest names based on certain criteria.
  Playing games: I can play simple games like Hangman, Trivia, and more.
  Providing recommendations: I can recommend movies, music, books, and other forms of entertainment based on user preferences.
  These are just some of the many functions that I am capable of performing. If you have a specific task in mind, feel free to ask me about it.
User: Tell me a joke.
Orion-14B: Sure, here's a classic one-liner: Why don't scientists trust atoms? Because they make up everything.

4.4.2. Japanese & Korean Chat

User:자기소개를 해주세요.
Orion-14B:안녕하세요, 제 이름은 ChatMax입니다. OrionStar에서 개발한 AI 어시스턴트입니다. 저는 사용자들에게 정확하고 객관적이며 유용한 답변을 제공하여 문제를 해결하고 정보를 얻는 데 도움을 줄 수 있습니다.

5.1. Declarations

We strongly urge all users not to use the Orion-14B model for any activities that may harm national or social security or violate the law.
Additionally, we request users not to use the Orion-14B model for internet services without proper security review and filing.
We hope all users abide by this principle to ensure that technological development takes place in a regulated and legal environment.
We have done our best to ensure the compliance of the data used in the model training process. However, despite our
significant efforts, unforeseen issues may still arise due to the complexity of the model and data. Therefore, if any
problems arise due to the use of the Orion-14B open-source model, including but not limited to data security
issues, public opinion risks, or any risks and issues arising from the model being misled, abused, disseminated, or
improperly utilized, we will not assume any responsibility.

5.2. License

Community use of the Orion-14B series models

OrionStar is a leading global service robot solutions company, founded in September 2016. OrionStar is dedicated to
using artificial intelligence technology to create the next generation of revolutionary robots, allowing people to break
free from repetitive physical labor and making human work and life more intelligent and enjoyable. Through technology,
OrionStar aims to make society and the world a better place.

OrionStar possesses fully self-developed end-to-end artificial intelligence technologies, such as voice interaction and
visual navigation. It integrates product development capabilities and technological application capabilities. Based on
the Orion robotic arm platform, it has launched products such as OrionStar AI Robot Greeting, AI Robot Greeting Mini,
Lucki, Coffee Master, and established the open platform OrionOS for Orion robots. Following the philosophy of “Born for
Truly Useful Robots”, OrionStar empowers more people through AI technology.

The core strengths of OrionStar lies in possessing end-to-end AI application capabilities, including big data preprocessing, large model pretraining, fine-tuning, prompt engineering, agent, etc. With comprehensive end-to-end model training capabilities, including systematic data processing workflows and the parallel model training capability of hundreds of GPUs, it has been successfully applied in various industry scenarios such as government affairs, cloud services, international e-commerce, and fast-moving consumer goods.

Companies with demands for deploying large-scale model applications are welcome to contact us.
Enquiry Hotline: 400-898-7779
E-mail: ai@orionstar.com


The Orion-14B series models by OrionStarAI have a wide range of use cases across various industries and scenarios. Some of the prominent use cases of these models include:

1. Language Understanding: The models show excellent performance in understanding and processing languages such as Chinese, English, Japanese, and Korean. They can be used for tasks such as language translation, comprehension, and generation of text in multiple languages.

2. Chatbot Applications: The Orion-14B-Chat model is fine-tuned to provide an interactive and engaging chat experience for users. It can be deployed as a chatbot for customer service, information retrieval, and conversation-based applications.

3. Long-Form Text Processing: The Orion-14B-LongChat model excels at handling extremely lengthy texts, making it suitable for applications such as summarization, narrative processing, and multi-field question-answering.

4. Multilingual Applications: Due to their strong multilingual capabilities, these models are ideal for developing applications that require support for multiple languages, such as multilingual search engines, language learning platforms, and global customer support systems.

5. Retrieval Augmented Generation: The Orion-14B-Chat-RAG model is designed for tasks that involve retrieving and generating text-based on specific queries, making it suitable for applications in information retrieval, question-answering systems, and intelligent assistants.

6. Quantized Model Applications: The Orion-14B-Base-Int4, Orion-14B-Chat-Int4, and other quantized models reduce the model size, improve inference speed, and are suitable for deployment on resource-constrained devices such as mobile phones, IoT devices, and edge computing systems.

7. Conversational AI: The models can be used to develop conversational AI applications, virtual assistants, and dialogue systems for use in various domains including healthcare, finance, education, and entertainment.

8. Intent Recognition and Plugin Tasks: The Orion-14B-Chat-Plugin model is tailored for intent recognition and function calling tasks, making it suitable for agent-related scenarios where the model acts as a plugin and function call system.

9. Model Inference: The models can be used for various inference tasks, including text generation, question-answering, summarization, and dialogue systems, and can be employed in software applications, APIs, and command-line tools.

10. Compliance and Ethical Implementation: It is essential to use these models responsibly and ethically, ensuring that they are not used for activities that may harm national or social security or violate the law. Proper security review and legal compliance are crucial for the internet services powered by these models.

As the Orion-14B series models are open-source and available for community use, they offer a wide array of possibilities for developers, researchers, and organizations across different sectors to leverage the power of large language models for innovative and impactful applications.