Turtle Benchmark is a new, cheat-proof benchmark based on the 'Turtle Soup' game, focusing on the assessment of large language models (LLMs) in terms of logical reasoning and context comprehension. By eliminating the need for background knowledge, it provides objective and unbiased test results with quantifiable outcomes, ensuring that models cannot be 'gamed' through the use of real user-generated questions.

The Prompt Report is a comprehensive systematic review report focusing on prompting techniques in generative artificial intelligence (GenAI). Through a collaborative effort of humans and machines, it processed 4797 records from multiple databases, extracting 1565 relevant papers. The report presents 58 text-based prompting techniques and complements them with a wide collection of multimodal and multilingual techniques. Its goal is to offer an accessible and implementable directory of prompting techniques and reviews agent-based prompt extensions, including evaluating outputs and designing prompts that promote safety and robustness. Additionally, the report demonstrates the application of prompting techniques in two case studies.

Open LLM Leaderboard, hosted by Hugging Face, is a space designed to showcase and compare the performance of various large language models. It offers a platform for developers, researchers, and enterprises to view the performance of different models on specific tasks, ultimately helping users choose the model best suited to their needs.

MMStar is a benchmark dataset designed to assess the multimodal capabilities of large visual language models. It comprises 1500 carefully selected visual language samples, covering 6 core abilities and 18 sub-dimensions. Each sample has undergone human review, ensuring visual dependency, minimizing data leakage, and requiring advanced multimodal capabilities for resolution. In addition to traditional accuracy metrics, MMStar proposes two new metrics to measure data leakage and the practical performance gains of multimodal training. Researchers can use MMStar to evaluate the multimodal capabilities of visual language models across multiple tasks and leverage the new metrics to discover potential issues within models.

This tool aims to assess the generalization ability, trustworthiness, and causal reasoning abilities of the latest proprietary and open-source MLLMs through qualitative research from four modalities: text, code, images, and videos. This is done to increase the transparency of MLLMs. We believe these attributes are representative factors defining the reliability of MLLMs, supporting various downstream applications. Specifically, we evaluated closed-source GPT-4 and Gemini, as well as 6 open-source LLMs and MLLMs. Overall, we evaluated 230 manually designed cases, with qualitative results summarized into 12 scores (i.e., 4 modalities multiplied by 3 attributes). In total, we revealed 14 empirical findings that contribute to understanding the capabilities and limitations of proprietary and open-source MLLMs, enabling more reliable support for multi-modal downstream applications.

GPTEval3D is an open-source tool for evaluating 3D generation models. Based on GPT-4V, it enables automatic evaluation of text-to-3D generation models. It can calculate the ELO score of the generated models and compare them with existing models for ranking. This user-friendly tool supports custom evaluation datasets, allowing users to fully leverage the evaluation capabilities of GPT-4V. It serves as a powerful tool for researching 3D generation tasks.

Deepmark AI is a benchmark tool for evaluating large language models (LLMs) that allows you to assess a variety of task-specific metrics on your own data. It comes pre-integrated with leading generative AI APIs like GPT-4, Anthropic, GPT-3.5 Turbo, Cohere, and AI21.

DeepEval provides a range of metrics to assess the quality of LLM's answers to ensure they are relevant, consistent, unbiased, and non-toxic. These can be easily integrated into CI/CD pipelines, enabling machine learning engineers to quickly assess and verify the performance of their LLM applications during iterative improvements. DeepEval offers a Python-friendly offline evaluation method, ensuring your pipeline is ready for production. It's like 'Pytest for your pipeline', making the process of production and evaluation as straightforward as passing all tests.

MLE-bench is a benchmark test launched by OpenAI to measure the performance of AI agents in the domain of machine learning engineering. It compiles 75 diverse challenges from Kaggle-related machine learning engineering competitions, testing real-world skills such as model training, dataset preparation, and experiment execution. Using publicly available leaderboard data from Kaggle, human benchmarks for each competition are established. Various cutting-edge language models are evaluated against this benchmark using open-source agent frameworks, revealing that the best-performing setup—OpenAI's o1-preview paired with the AIDE framework—achieved at least Kaggle bronze medal levels in 16.9% of the competitions. Moreover, the study examines various resource extension forms of AI agents and the effects of pre-training contamination. The benchmark code for MLE-bench has been open-sourced to facilitate future understanding of AI agents' capabilities in machine learning engineering.

SWE-bench Verified is a subset of SWE-bench released by OpenAI that has been manually verified to reliably assess the ability of AI models to solve real-world software issues. It challenges AI to generate patches that resolve the described problems by providing code repositories and problem descriptions. This tool has been developed to improve the accuracy of evaluating the model's ability to autonomously perform software engineering tasks and is a key component of OpenAI's medium-risk framework.

Turtle Benchmark is a new, cheat-proof benchmark based on the 'Turtle Soup' game, focusing on the assessment of large language models (LLMs) in terms of logical reasoning and context comprehension. By eliminating the need for background knowledge, it provides objective and unbiased test results with quantifiable outcomes, ensuring that models cannot be 'gamed' through the use of real user-generated questions.

The Prompt Report is a comprehensive systematic review report focusing on prompting techniques in generative artificial intelligence (GenAI). Through a collaborative effort of humans and machines, it processed 4797 records from multiple databases, extracting 1565 relevant papers. The report presents 58 text-based prompting techniques and complements them with a wide collection of multimodal and multilingual techniques. Its goal is to offer an accessible and implementable directory of prompting techniques and reviews agent-based prompt extensions, including evaluating outputs and designing prompts that promote safety and robustness. Additionally, the report demonstrates the application of prompting techniques in two case studies.

Open LLM Leaderboard, hosted by Hugging Face, is a space designed to showcase and compare the performance of various large language models. It offers a platform for developers, researchers, and enterprises to view the performance of different models on specific tasks, ultimately helping users choose the model best suited to their needs.

OpenCompass 2.0 is a platform dedicated to evaluating the performance of large language models. It utilizes multiple closed-source datasets for multi-dimensional assessments, providing models with an overall average score and specialized skill scores. The platform helps developers and researchers understand the performance of different models in areas like language, knowledge, reasoning, mathematics, and programming through its real-time updated leaderboard.

MMStar is a benchmark dataset designed to assess the multimodal capabilities of large visual language models. It comprises 1500 carefully selected visual language samples, covering 6 core abilities and 18 sub-dimensions. Each sample has undergone human review, ensuring visual dependency, minimizing data leakage, and requiring advanced multimodal capabilities for resolution. In addition to traditional accuracy metrics, MMStar proposes two new metrics to measure data leakage and the practical performance gains of multimodal training. Researchers can use MMStar to evaluate the multimodal capabilities of visual language models across multiple tasks and leverage the new metrics to discover potential issues within models.

This tool aims to assess the generalization ability, trustworthiness, and causal reasoning abilities of the latest proprietary and open-source MLLMs through qualitative research from four modalities: text, code, images, and videos. This is done to increase the transparency of MLLMs. We believe these attributes are representative factors defining the reliability of MLLMs, supporting various downstream applications. Specifically, we evaluated closed-source GPT-4 and Gemini, as well as 6 open-source LLMs and MLLMs. Overall, we evaluated 230 manually designed cases, with qualitative results summarized into 12 scores (i.e., 4 modalities multiplied by 3 attributes). In total, we revealed 14 empirical findings that contribute to understanding the capabilities and limitations of proprietary and open-source MLLMs, enabling more reliable support for multi-modal downstream applications.