
mint-bench
Official Repo for ICLR 2024 paper MINT: Evaluating LLMs in Multi-turn Interaction with Tools and Language Feedback by Xingyao Wang*, Zihan Wang*, Jiateng Liu, Yangyi Chen, Lifan Yuan, Hao Peng and Heng Ji.
Stars: 84

MINT benchmark aims to evaluate LLMs' ability to solve tasks with multi-turn interactions by (1) using tools and (2) leveraging natural language feedback.
README:
Official Repo for paper MINT: Evaluating LLMs in Multi-turn Interaction with Tools and Language Feedback by Xingyao Wang*, Zihan Wang*, Jiateng Liu, Yangyi Chen, Lifan Yuan, Hao Peng and Heng Ji.
MINT benchmark aims to evaluate LLMs' ability to solve tasks with multi-turn interactions by (1) using tools and (2) leveraging natural language feedback.
๐ Please visit our website for the leaderboard.
โ ๏ธ WARNING: Evaluation of LLMs requires executing untrusted model-generated code. Users are strongly encouraged to sandbox the code execution so that it does not perform destructive actions on their host or network. We highly recommend using the provided docker image for isolated execution.
git clone https://github.com/xingyaoww/mint-bench
cd mint-bench
You can choose to use docker (recommended) or local setup as follows.
You only need to ensure that you have docker installed on your local computer following the official guide.
# Install Anaconda or Miniconda first if you don't have it already.
conda env create -f environment.yml
conda activate mint
# Install the MINT package
pip install -e .
To evaluate huggingface-compatible open-source models, check instructions here.
To evaluate API-base closed-source models, set Your API Key:
# Obtain OpenAI API access from https://openai.com/blog/openai-api
# This key is necessary for all models since, by default, we use GPT-4 for feedback generation
export OPENAI_API_KEY='YOUR KEY HERE';
# The following keys are optional
# Will only be used if you use Bard or Claude as the evaluated model/feedback provider
# https://www.googlecloudcommunity.com/gc/AI-ML/Google-Bard-API/m-p/538517
export BARD_API_KEY='YOUR KEY HERE';
# https://docs.anthropic.com/claude/docs/getting-access-to-claude
export ANTHROPIC_API_KEY='YOUR KEY HERE';
Please refer to docs/CONFIG.md if you want to evaluate a custom API-based model.
MINT uses configuration files to specify the experiment settings (e.g., which models to use, which datasets to use, etc.).
We provide a script mint/configs/config_variables.py
to generate default config files for you.
Please refer to docs/CONFIG.md if you want to add an evaluated model, a different LLM that provides feedback, or different experiment setting (e.g., feedback type).
# Create configuration files for your experiment.
# By default configs will be saved to configs/ and model outputs will be saved to data/outputs
python mint/configs/generate_config.py
If you want to use a docker environment for better isolation: We recommend using a terminal multiplexer (e.g., tmux) to start the docker for better management. Please make sure you have admin permission by adding the prefix "sudo" or have your admin add you to the docker user group. You will enter an interactive docker environment with all dependencies installed:
./scripts/docker/run_interactive.sh
Run the following once you are inside the docker container (for docker-setup), or you are in the conda environment (conda activate mint
):
# Use this script to run the experiment
# You can specify which config file you want to run by modifying the script
# By default, we will run multiple configs in parallel and save output to `output.txt` in corresponding outputs folder
./scripts/run.sh
# If you want to run the experiment in debug mode (only run 1 config at a time), run the following
DEBUG_MODE=1 ./scripts/run.sh
Analyze result: You can use the notebook scripts/notebook/analyze_output.ipynb
to analyze the model you evaluated and reproduce tables and figures from the paper.
Alternatively, you can convert output by running python scripts/convert_outputs.py --data_dir data/outputs --output_dir YOUR_OUTPUT_DIR/
to get results breakdown by task name.
Visualize the results: The following code will starts a streamlit-based visualizer to visualize your directory.
streamlit run scripts/visualizer.py -- --data_dir data/outputs
We welcome contributions to this repo from the community!
To add your model to the benchmark, you can start a PR to merge your model outputs data/outputs
into this repo. This will be automatically updated to the website.
Please refer to docs/CONTRIBUTING.md for more details about contributing models, tools, and data.
If you find this repo helpful, please cite our paper:
@misc{wang2023mint,
title={MINT: Evaluating LLMs in Multi-turn Interaction with Tools and Language Feedback},
author={Xingyao Wang and Zihan Wang and Jiateng Liu and Yangyi Chen and Lifan Yuan and Hao Peng and Heng Ji},
year={2023},
eprint={2309.10691},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
For Tasks:
Click tags to check more tools for each tasksFor Jobs:
Alternative AI tools for mint-bench
Similar Open Source Tools

mint-bench
MINT benchmark aims to evaluate LLMs' ability to solve tasks with multi-turn interactions by (1) using tools and (2) leveraging natural language feedback.

llm-applications
A comprehensive guide to building Retrieval Augmented Generation (RAG)-based LLM applications for production. This guide covers developing a RAG-based LLM application from scratch, scaling the major components, evaluating different configurations, implementing LLM hybrid routing, serving the application in a highly scalable and available manner, and sharing the impacts LLM applications have had on products.

telemetry-airflow
This repository codifies the Airflow cluster that is deployed at workflow.telemetry.mozilla.org (behind SSO) and commonly referred to as "WTMO" or simply "Airflow". Some links relevant to users and developers of WTMO: * The `dags` directory in this repository contains some custom DAG definitions * Many of the DAGs registered with WTMO don't live in this repository, but are instead generated from ETL task definitions in bigquery-etl * The Data SRE team maintains a WTMO Developer Guide (behind SSO)

DesktopCommanderMCP
Desktop Commander MCP is a server that allows the Claude desktop app to execute long-running terminal commands on your computer and manage processes through Model Context Protocol (MCP). It is built on top of MCP Filesystem Server to provide additional search and replace file editing capabilities. The tool enables users to execute terminal commands with output streaming, manage processes, perform full filesystem operations, and edit code with surgical text replacements or full file rewrites. It also supports vscode-ripgrep based recursive code or text search in folders.

CoolCline
CoolCline is a proactive programming assistant that combines the best features of Cline, Roo Code, and Bao Cline. It seamlessly collaborates with your command line interface and editor, providing the most powerful AI development experience. It optimizes queries, allows quick switching of LLM Providers, and offers auto-approve options for actions. Users can configure LLM Providers, select different chat modes, perform file and editor operations, integrate with the command line, automate browser tasks, and extend capabilities through the Model Context Protocol (MCP). Context mentions help provide explicit context, and installation is easy through the editor's extension panel or by dragging and dropping the `.vsix` file. Local setup and development instructions are available for contributors.

neural
Neural is a Vim and Neovim plugin that integrates various machine learning tools to assist users in writing code, generating text, and explaining code or paragraphs. It supports multiple machine learning models, focuses on privacy, and is compatible with Vim 8.0+ and Neovim 0.8+. Users can easily configure Neural to interact with third-party machine learning tools, such as OpenAI, to enhance code generation and completion. The plugin also provides commands like `:NeuralExplain` to explain code or text and `:NeuralStop` to stop Neural from working. Neural is maintained by the Dense Analysis team and comes with a disclaimer about sending input data to third-party servers for machine learning queries.

talon-ai-tools
Control large language models and AI tools through voice commands using the Talon Voice dictation engine. This tool is designed to help users quickly edit text, code by voice, reduce keyboard use for those with health issues, and speed up workflow by using AI commands across the desktop. It prompts and extends tools like Github Copilot and OpenAI API for text and image generation. Users can set up the tool by downloading the repo, obtaining an OpenAI API key, and customizing the endpoint URL for preferred models. The tool can be used without an OpenAI key and can be exclusively used with Copilot for those not needing LLM integration.

AgentIQ
AgentIQ is a flexible library designed to seamlessly integrate enterprise agents with various data sources and tools. It enables true composability by treating agents, tools, and workflows as simple function calls. With features like framework agnosticism, reusability, rapid development, profiling, observability, evaluation system, user interface, and MCP compatibility, AgentIQ empowers developers to move quickly, experiment freely, and ensure reliability across agent-driven projects.

AI-Scientist
The AI Scientist is a comprehensive system for fully automatic scientific discovery, enabling Foundation Models to perform research independently. It aims to tackle the grand challenge of developing agents capable of conducting scientific research and discovering new knowledge. The tool generates papers on various topics using Large Language Models (LLMs) and provides a platform for exploring new research ideas. Users can create their own templates for specific areas of study and run experiments to generate papers. However, caution is advised as the codebase executes LLM-written code, which may pose risks such as the use of potentially dangerous packages and web access.

patchwork
PatchWork is an open-source framework designed for automating development tasks using large language models. It enables users to automate workflows such as PR reviews, bug fixing, security patching, and more through a self-hosted CLI agent and preferred LLMs. The framework consists of reusable atomic actions called Steps, customizable LLM prompts known as Prompt Templates, and LLM-assisted automations called Patchflows. Users can run Patchflows locally in their CLI/IDE or as part of CI/CD pipelines. PatchWork offers predefined patchflows like AutoFix, PRReview, GenerateREADME, DependencyUpgrade, and ResolveIssue, with the flexibility to create custom patchflows. Prompt templates are used to pass queries to LLMs and can be customized. Contributions to new patchflows, steps, and the core framework are encouraged, with chat assistants available to aid in the process. The roadmap includes expanding the patchflow library, introducing a debugger and validation module, supporting large-scale code embeddings, parallelization, fine-tuned models, and an open-source GUI. PatchWork is licensed under AGPL-3.0 terms, while custom patchflows and steps can be shared using the Apache-2.0 licensed patchwork template repository.

srcbook
Srcbook is an open-source interactive programming environment for TypeScript that allows users to create, run, and share reproducible programs and ideas. It features AI capabilities for exploring and iterating on ideas, supports exporting to valid markdown format, and enables diagraming with mermaid for rich annotations. Users can locally execute programs through a web interface, powered by Node.js under the Apache2 license.

node_characterai
Node.js client for the unofficial Character AI API, an awesome website which brings characters to life with AI! This repository is inspired by RichardDorian's unofficial node API. Though, I found it hard to use and it was not really stable and archived. So I remade it in javascript. This project is not affiliated with Character AI in any way! It is a community project. The purpose of this project is to bring and build projects powered by Character AI. If you like this project, please check their website.

torchchat
torchchat is a codebase showcasing the ability to run large language models (LLMs) seamlessly. It allows running LLMs using Python in various environments such as desktop, server, iOS, and Android. The tool supports running models via PyTorch, chatting, generating text, running chat in the browser, and running models on desktop/server without Python. It also provides features like AOT Inductor for faster execution, running in C++ using the runner, and deploying and running on iOS and Android. The tool supports popular hardware and OS including Linux, Mac OS, Android, and iOS, with various data types and execution modes available.

GraphRAG-Local-UI
GraphRAG Local with Interactive UI is an adaptation of Microsoft's GraphRAG, tailored to support local models and featuring a comprehensive interactive user interface. It allows users to leverage local models for LLM and embeddings, visualize knowledge graphs in 2D or 3D, manage files, settings, and queries, and explore indexing outputs. The tool aims to be cost-effective by eliminating dependency on costly cloud-based models and offers flexible querying options for global, local, and direct chat queries.

qrev
QRev is an open-source alternative to Salesforce, offering AI agents to scale sales organizations infinitely. It aims to provide digital workers for various sales roles or a superagent named Qai. The tech stack includes TypeScript for frontend, NodeJS for backend, MongoDB for app server database, ChromaDB for vector database, SQLite for AI server SQL relational database, and Langchain for LLM tooling. The tool allows users to run client app, app server, and AI server components. It requires Node.js and MongoDB to be installed, and provides detailed setup instructions in the README file.

markdowner
Markdowner is a fast tool designed to convert any website into LLM-ready markdown data. It aims to improve the quality of responses in the AI app Supermemory by structuring and predicting data in markdown format. The tool offers features such as website conversion, LLM filtering, detailed markdown mode, auto crawler, text and JSON responses, and easy self-hosting. Markdowner utilizes Cloudflare's Browser rendering and Durable objects for browser instance creation and markdown conversion. Users can self-host the project with the Workers paid plan, following simple steps. Support the project by starring the repository.
For similar tasks

LLMStack
LLMStack is a no-code platform for building generative AI agents, workflows, and chatbots. It allows users to connect their own data, internal tools, and GPT-powered models without any coding experience. LLMStack can be deployed to the cloud or on-premise and can be accessed via HTTP API or triggered from Slack or Discord.

ai-guide
This guide is dedicated to Large Language Models (LLMs) that you can run on your home computer. It assumes your PC is a lower-end, non-gaming setup.

onnxruntime-genai
ONNX Runtime Generative AI is a library that provides the generative AI loop for ONNX models, including inference with ONNX Runtime, logits processing, search and sampling, and KV cache management. Users can call a high level `generate()` method, or run each iteration of the model in a loop. It supports greedy/beam search and TopP, TopK sampling to generate token sequences, has built in logits processing like repetition penalties, and allows for easy custom scoring.

jupyter-ai
Jupyter AI connects generative AI with Jupyter notebooks. It provides a user-friendly and powerful way to explore generative AI models in notebooks and improve your productivity in JupyterLab and the Jupyter Notebook. Specifically, Jupyter AI offers: * An `%%ai` magic that turns the Jupyter notebook into a reproducible generative AI playground. This works anywhere the IPython kernel runs (JupyterLab, Jupyter Notebook, Google Colab, Kaggle, VSCode, etc.). * A native chat UI in JupyterLab that enables you to work with generative AI as a conversational assistant. * Support for a wide range of generative model providers, including AI21, Anthropic, AWS, Cohere, Gemini, Hugging Face, NVIDIA, and OpenAI. * Local model support through GPT4All, enabling use of generative AI models on consumer grade machines with ease and privacy.

khoj
Khoj is an open-source, personal AI assistant that extends your capabilities by creating always-available AI agents. You can share your notes and documents to extend your digital brain, and your AI agents have access to the internet, allowing you to incorporate real-time information. Khoj is accessible on Desktop, Emacs, Obsidian, Web, and Whatsapp, and you can share PDF, markdown, org-mode, notion files, and GitHub repositories. You'll get fast, accurate semantic search on top of your docs, and your agents can create deeply personal images and understand your speech. Khoj is self-hostable and always will be.

langchain_dart
LangChain.dart is a Dart port of the popular LangChain Python framework created by Harrison Chase. LangChain provides a set of ready-to-use components for working with language models and a standard interface for chaining them together to formulate more advanced use cases (e.g. chatbots, Q&A with RAG, agents, summarization, extraction, etc.). The components can be grouped into a few core modules: * **Model I/O:** LangChain offers a unified API for interacting with various LLM providers (e.g. OpenAI, Google, Mistral, Ollama, etc.), allowing developers to switch between them with ease. Additionally, it provides tools for managing model inputs (prompt templates and example selectors) and parsing the resulting model outputs (output parsers). * **Retrieval:** assists in loading user data (via document loaders), transforming it (with text splitters), extracting its meaning (using embedding models), storing (in vector stores) and retrieving it (through retrievers) so that it can be used to ground the model's responses (i.e. Retrieval-Augmented Generation or RAG). * **Agents:** "bots" that leverage LLMs to make informed decisions about which available tools (such as web search, calculators, database lookup, etc.) to use to accomplish the designated task. The different components can be composed together using the LangChain Expression Language (LCEL).

danswer
Danswer is an open-source Gen-AI Chat and Unified Search tool that connects to your company's docs, apps, and people. It provides a Chat interface and plugs into any LLM of your choice. Danswer can be deployed anywhere and for any scale - on a laptop, on-premise, or to cloud. Since you own the deployment, your user data and chats are fully in your own control. Danswer is MIT licensed and designed to be modular and easily extensible. The system also comes fully ready for production usage with user authentication, role management (admin/basic users), chat persistence, and a UI for configuring Personas (AI Assistants) and their Prompts. Danswer also serves as a Unified Search across all common workplace tools such as Slack, Google Drive, Confluence, etc. By combining LLMs and team specific knowledge, Danswer becomes a subject matter expert for the team. Imagine ChatGPT if it had access to your team's unique knowledge! It enables questions such as "A customer wants feature X, is this already supported?" or "Where's the pull request for feature Y?"

infinity
Infinity is an AI-native database designed for LLM applications, providing incredibly fast full-text and vector search capabilities. It supports a wide range of data types, including vectors, full-text, and structured data, and offers a fused search feature that combines multiple embeddings and full text. Infinity is easy to use, with an intuitive Python API and a single-binary architecture that simplifies deployment. It achieves high performance, with 0.1 milliseconds query latency on million-scale vector datasets and up to 15K QPS.
For similar jobs

weave
Weave is a toolkit for developing Generative AI applications, built by Weights & Biases. With Weave, you can log and debug language model inputs, outputs, and traces; build rigorous, apples-to-apples evaluations for language model use cases; and organize all the information generated across the LLM workflow, from experimentation to evaluations to production. Weave aims to bring rigor, best-practices, and composability to the inherently experimental process of developing Generative AI software, without introducing cognitive overhead.

agentcloud
AgentCloud is an open-source platform that enables companies to build and deploy private LLM chat apps, empowering teams to securely interact with their data. It comprises three main components: Agent Backend, Webapp, and Vector Proxy. To run this project locally, clone the repository, install Docker, and start the services. The project is licensed under the GNU Affero General Public License, version 3 only. Contributions and feedback are welcome from the community.

oss-fuzz-gen
This framework generates fuzz targets for real-world `C`/`C++` projects with various Large Language Models (LLM) and benchmarks them via the `OSS-Fuzz` platform. It manages to successfully leverage LLMs to generate valid fuzz targets (which generate non-zero coverage increase) for 160 C/C++ projects. The maximum line coverage increase is 29% from the existing human-written targets.

LLMStack
LLMStack is a no-code platform for building generative AI agents, workflows, and chatbots. It allows users to connect their own data, internal tools, and GPT-powered models without any coding experience. LLMStack can be deployed to the cloud or on-premise and can be accessed via HTTP API or triggered from Slack or Discord.

VisionCraft
The VisionCraft API is a free API for using over 100 different AI models. From images to sound.

kaito
Kaito is an operator that automates the AI/ML inference model deployment in a Kubernetes cluster. It manages large model files using container images, avoids tuning deployment parameters to fit GPU hardware by providing preset configurations, auto-provisions GPU nodes based on model requirements, and hosts large model images in the public Microsoft Container Registry (MCR) if the license allows. Using Kaito, the workflow of onboarding large AI inference models in Kubernetes is largely simplified.

PyRIT
PyRIT is an open access automation framework designed to empower security professionals and ML engineers to red team foundation models and their applications. It automates AI Red Teaming tasks to allow operators to focus on more complicated and time-consuming tasks and can also identify security harms such as misuse (e.g., malware generation, jailbreaking), and privacy harms (e.g., identity theft). The goal is to allow researchers to have a baseline of how well their model and entire inference pipeline is doing against different harm categories and to be able to compare that baseline to future iterations of their model. This allows them to have empirical data on how well their model is doing today, and detect any degradation of performance based on future improvements.

Azure-Analytics-and-AI-Engagement
The Azure-Analytics-and-AI-Engagement repository provides packaged Industry Scenario DREAM Demos with ARM templates (Containing a demo web application, Power BI reports, Synapse resources, AML Notebooks etc.) that can be deployed in a customerโs subscription using the CAPE tool within a matter of few hours. Partners can also deploy DREAM Demos in their own subscriptions using DPoC.