Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
RankLLM is a Python toolkit for reproducible information retrieval research using rerankers, with a focus on listwise reranking.
| Date | Stars |
|---|---|
| 2026-07-31 | 611 |
| 2026-08-03 | 610 |
| 2026-08-06 | 610 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# RankLLM [](https://pypi.org/project/rank-llm/) [](https://pepy.tech/project/rank-llm) [](https://pepy.tech/project/rank-llm) [](https://arxiv.org/abs/2309.15088) [](https://www.apache.org/licenses/LICENSE-2.0) ## News - **[2026.03.26]** RankLLM now supports the new `rank-llm` command-line interface (CLI). - **[2025.08.25]** Added support for OpenRouter API - Release [v0.25.7](docs/release-notes/release-notes-v0.25.7.md) - **[2025.07.23]** Added support for custom prompt templates with YAML files - Release [v0.25.0](docs/release-notes/release-notes-v0.25.0.md). You can now integrate your own prompt and language model with just a few lines of code. Checkout the [Reasonrank integration](https://github.com/castorini/rank_llm/pull/306) as an example. - **[2025.05.25]** Our [RankLLM](https://dl.acm.org/doi/pdf/10.1145/3726302.3730331) resource paper is accepted to SIGIR 2025! 🎉🎉🎉 ## Overview We offer a suite of rerankers - pointwise models like MonoT5, pairwise models like DuoT5 and listwise models with a focus on open source LLMs compatible with [vLLM](https://https://github.com/vllm-project/vllm), [SGLang](https://github.com/sgl-project/sglang). We also support RankGPT and RankGemini variants, which are proprietary listwise rerankers. Addtionally, we support reranking with the first-token logits only to improve inference efficiency. Some of the code in this repository is borrowed from [RankGPT](https://github.com/sunnweiwei/RankGPT), [PyGaggle](https://github.com/castorini/pygaggle), and [LiT5](https://github.com/castorini/LiT5)! <p align="center"> <img src="docs/images/rankllm_overview.png" alt="RankLLM Overview" style="width:95%;"> </p> ## Releases current_version = "0.25.7" ## Content 1. [Installation](#installation) 2. [Quick Start](#quick-start) 3. [End-to-end Run and 2CR](#end-to-end-run-and-2cr) 4. [Model Zoo](#model-zoo) 5. [Training](#training) 6. [Community Contribution](#community-contribution) 7. [References and Citations](#references) 8. [Acknowledgments](#acknowledgments) <a id="installation"></a> # 📟 Installation `uv` is the canonical contributor workflow for this repository. The existing `conda` and `pip` paths remain available as fallbacks. ## Install `uv` Install `uv` with Astral's official installer: ```bash curl -LsSf https://astral.sh/uv/install.sh | sh export PATH="$HOME/.local/bin:$PATH" ``` ## Prerequisites - Install Java 21 only if you plan to use retrieval or evaluation workflows via `rank-llm[pyserini]`. JDK 11 is not supported. - Install CUDA-specific PyTorch wheels separately if you want GPU-optimized builds beyond the default Python package resolution. ## Development Installation For development or the latest features, create a repo-local virtual environment: ```bash git clone https://github.com/castorini/rank_llm.git cd rank_llm uv python install 3.11 uv venv --python 3.11 source .venv/bin/activate uv sync --group dev ``` If you prefer not to activate the virtual environment, run commands through `uv run`, for example `uv run python -m unittest discover test`. ## Optional Extras Install only the stacks you need: ```bash uv sync --group dev --extra <extra> ``` Replace `<extra>` with one of the extras in the feature matrix below. You can repeat `--extra` to combine stacks in one environment. ### Feature Matrix | Workflow | Extra | Notes | | --- | --- | --- | | Hosted OpenAI or OpenRouter rerankers | `openai` | Includes `python-doten
Excerpt of 25,251 characters
Read on GitHub
86
45
40
10
9
7
6
6
5
4
4
3
3
3
Brayden Zhong · @radixark
2
Jasper Xian
2
2
2
2
2
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:b1c7b8cb57f8734b, desc:reproducible