👋 join us on Discord and WeChat
We are honored to have witnessed the tremendous progress of artificial general intelligence together with the community in the past year, and we are also very pleased that OpenCompass can help numerous developers and users.
We announce the launch of the OpenCompass 2023 LLM Annual Leaderboard plan. We expect to release the annual leaderboard of the LLMs in January 2024, systematically evaluating the performance of LLMs in various capabilities such as language, knowledge, reasoning, creation, long-text, and agents.
At that time, we will release rankings for both open-source models and commercial API models, aiming to provide a comprehensive, objective, and neutral reference for the industry and research community.
We sincerely invite various large models to join the OpenCompass to showcase their performance advantages in different fields. At the same time, we also welcome researchers and developers to provide valuable suggestions and contributions to jointly promote the development of the LLMs. If you have any questions or needs, please feel free to contact us. In addition, relevant evaluation contents, performance statistics, and evaluation methods will be open-source along with the leaderboard release.
We have provided the more details of the CompassBench 2023 in Doc.
Let's look forward to the release of the OpenCompass 2023 LLM Annual Leaderboard!
to OpenCompass!
Just like a compass guides us on our journey, OpenCompass will guide you through the complex landscape of evaluating large language models. With its powerful algorithms and intuitive interface, OpenCompass makes it easy to assess the quality and effectiveness of your NLP models.
🚩🚩🚩 Explore opportunities at OpenCompass! We're currently hiring full-time researchers/engineers and interns. If you're passionate about LLM and OpenCompass, don't hesitate to reach out to us via email. We'd love to hear from you!
🔥🔥🔥 We are delighted to announce that the OpenCompass has been recommended by the Meta AI, click Get Started of Llama for more information.
Attention
We launch the OpenCompass Collaboration project, welcome to support diverse evaluation benchmarks into OpenCompass! Clike Issue for more information. Let's work together to build a more powerful OpenCompass toolkit!
- [2024.01.17] We supported the evaluation of InternLM2 and InternLM2-Chat, InternLM2 showed extremely strong performance in these tests, welcome to try! 🔥🔥🔥.
- [2024.01.17] We supported the needle in a haystack test with multiple needles, more information can be found here 🔥🔥🔥.
- [2023.12.28] We have enabled seamless evaluation of all models developed using LLaMA2-Accessory, a powerful toolkit for comprehensive LLM development. 🔥🔥🔥.
- [2023.12.22] We have released T-Eval, a step-by-step evaluation benchmark to gauge your LLMs on tool utilization. Welcome to our Leaderboard for more details! 🔥🔥🔥.
- [2023.12.10] We have released VLMEvalKit, a toolkit for evaluating vision-language models (VLMs), currently support 20+ VLMs and 7 multi-modal benchmarks (including MMBench series).
- [2023.12.10] We have supported Mistral AI's MoE LLM: Mixtral-8x7B-32K. Welcome to MixtralKit for more details about inference and evaluation.
OpenCompass is a one-stop platform for large model evaluation, aiming to provide a fair, open, and reproducible benchmark for large model evaluation. Its main features include:
-
Comprehensive support for models and datasets: Pre-support for 20+ HuggingFace and API models, a model evaluation scheme of 70+ datasets with about 400,000 questions, comprehensively evaluating the capabilities of the models in five dimensions.
-
Efficient distributed evaluation: One line command to implement task division and distributed evaluation, completing the full evaluation of billion-scale models in just a few hours.
-
Diversified evaluation paradigms: Support for zero-shot, few-shot, and chain-of-thought evaluations, combined with standard or dialogue-type prompt templates, to easily stimulate the maximum performance of various models.
-
Modular design with high extensibility: Want to add new models or datasets, customize an advanced task division strategy, or even support a new cluster management system? Everything about OpenCompass can be easily expanded!
-
Experiment management and reporting mechanism: Use config files to fully record each experiment, and support real-time reporting of results.
We provide OpenCompass Leaderboard for the community to rank all public models and API models. If you would like to join the evaluation, please provide the model repository URL or a standard API interface to the email address opencompass@pjlab.org.cn.
Below are the steps for quick installation and datasets preparation.
conda create --name opencompass python=3.10 pytorch torchvision pytorch-cuda -c nvidia -c pytorch -y
conda activate opencompass
git clone https://github.com/open-compass/opencompass opencompass
cd opencompass
pip install -e .conda create -n opencompass python=3.10 pytorch torchvision torchaudio cpuonly -c pytorch -y
conda activate opencompass
git clone https://github.com/open-compass/opencompass opencompass
cd opencompass
pip install -e .
# also please install requiresments packages via `pip install -r requirements/api.txt` for API models if needed.# Download dataset to data/ folder
wget https://github.com/open-compass/opencompass/releases/download/0.1.8.rc1/OpenCompassData-core-20231110.zip
unzip OpenCompassData-core-20231110.zipSome third-party features, like Humaneval and Llama, may require additional steps to work properly, for detailed steps please refer to the Installation Guide.
After ensuring that OpenCompass is installed correctly according to the above steps and the datasets are prepared, you can evaluate the performance of the LLaMA-7b model on the MMLU and C-Eval datasets using the following command:
python run.py --models hf_llama_7b --datasets mmlu_ppl ceval_pplOpenCompass has predefined configurations for many models and datasets. You can list all available model and dataset configurations using the tools.
# List all configurations
python tools/list_configs.py
# List all configurations related to llama and mmlu
python tools/list_configs.py llama mmluYou can also evaluate other HuggingFace models via command line. Taking LLaMA-7b as an example:
python run.py --datasets ceval_ppl mmlu_ppl \
--hf-path huggyllama/llama-7b \ # HuggingFace model path
--model-kwargs device_map='auto' \ # Arguments for model construction
--tokenizer-kwargs padding_side='left' truncation='left' use_fast=False \ # Arguments for tokenizer construction
--max-out-len 100 \ # Maximum number of tokens generated
--max-seq-len 2048 \ # Maximum sequence length the model can accept
--batch-size 8 \ # Batch size
--no-batch-padding \ # Don't enable batch padding, infer through for loop to avoid performance loss
--num-gpus 1 # Number of minimum required GPUsNote
To run the command above, you will need to remove the comments starting from#first.
Through the command line or configuration files, OpenCompass also supports evaluating APIs or custom models, as well as more diversified evaluation strategies. Please read the Quick Start to learn how to run an evaluation task.
| Language | Knowledge | Reasoning | Examination |
Word Definition
Idiom Learning
Semantic Similarity
Coreference Resolution
Translation
Multi-language Question Answering
Multi-language Summary
|
Knowledge Question Answering
|
Textual Entailment
Commonsense Reasoning
Mathematical Reasoning
Theorem Application
Comprehensive Reasoning
|
Junior High, High School, University, Professional Examinations
Medical Examinations
|
| Understanding | Long Context | Safety | Code |
Reading Comprehension
Content Summary
Content Analysis
|
Long Context Understanding
|
Safety
Robustness
|
Code
|
| Open-source Models | API Models |
|
- Subjective Evaluation
- Release CompassAreana
- Subjective evaluation dataset.
- Long-context
- Long-context evaluation with extensive datasets.
- Long-context leaderboard.
- Coding
- Coding evaluation leaderboard.
- Non-python language evaluation service.
- Agent
- Support various agenet framework.
- Evaluation of tool use of the LLMs.
- Robustness
- Support various attack method
We appreciate all contributions to improving OpenCompass. Please refer to the contributing guideline for the best practice.
Some code in this project is cited and modified from OpenICL.
Some datasets and prompt implementations are modified from chain-of-thought-hub and instruct-eval.
@misc{2023opencompass,
title={OpenCompass: A Universal Evaluation Platform for Foundation Models},
author={OpenCompass Contributors},
howpublished = {\url{https://github.com/open-compass/opencompass}},
year={2023}
}