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.
Multi-Faceted AI Agent and Workflow Autotuning. Automatically optimizes LangChain, LangGraph, DSPy programs for better quality, lower execution latency, and lower execution cost. Also has a simple agent/workflow framework
| Date | Stars |
|---|---|
| 2026-07-31 | 279 |
| 2026-08-01 | 280 |
| 2026-08-02 | 280 |
| 2026-08-06 | 280 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
<p align="center">
<img src="https://github.com/GenseeAI/cognify/blob/main/cognify.jpg?raw=true" alt="Cognify logo">
</p>
<p align="center">
| <a href="https://cognify-ai.readthedocs.io/en/latest/user_guide/quickstart.html"><b>Quickstart</b></a> | <a href="https://cognify-ai.readthedocs.io/en/latest/index.html"><b>Documentation</b></a> | <a href="https://arxiv.org/abs/2502.08056"><b>Research Paper</b></a> | <a href="https://www.gensee.ai/blog"><b>Blog</b></a> | <a href="https://discord.gg/8TSFeZA3V6"><b>Discord</b></a> | <a href="https://forms.gle/Be3MD3pGPpZaUmrVA"><b>Send Feedback</b></a> |
</p>
# Multi-Faceted AI Agent and Workflow Optimizer
Building high-quality, cost-effective gen-AI applications such as AI agents is challenging due to the absence of systematic methods for tuning, testing, and optimizing them.
We introduce **Cognify**, a tool that automatically enhances generation quality, decreases execution latency, and reduces execution monetary costs for AI agents and workflows, including those written with LangChain, LangGraph, DSPy, and Cognify's own framework.
Built on a novel foundation of hierarchical, workflow-level optimization, Cognify improves agent **generation quality by up to 2.8x**, reduces execution **cost
by up to 10x**, and reduces end-to-end **latency by up to 2.7x** compared to original expert-written agents.
## News
- <b>May 15, 2025</b>: 🎉Our paper has been accepted to **KDD 2025!** See you in Toronto this August!
- <b>Mar 13, 2025</b>: New blog series (Pt2) deep-diving into how Cognify achieves 2.8x quality improvement with just $5 and 24 minutes available [here](https://www.gensee.ai/post/auto-tuning-with-cognify-the-secret-to-boosting-your-gen-ai-workflow-quality-by-2-8-times-with-5-i-1)
- <b>Mar 12, 2025</b>: New blog series (Pt1) introducing how Cognify achieves 2.8x quality improvement with just $5 and 24 minutes available [here](https://www.gensee.ai/post/auto-tuning-with-cognify-the-secret-to-boosting-your-gen-ai-workflow-quality-by-2-8-times-with-5-i)
- <b>Feb 12, 2025</b>: Cognify [research paper](https://arxiv.org/abs/2502.08056) available on arxiv.
- <b>Nov 25, 2024</b>: Cognify announcement and overview blog post published [here](https://mlsys.wuklab.io/posts/cognify/).
## Installation
Cognify is available as a Python package and can be installed as
```
pip install cognify-ai
```
Or install from the source:
```
git clone https://github.com/GenseeAI/cognify
cd cognify
pip install -e .
```
## Getting Started
You can use Cognify with our simple CLI:
```bash
cognify optimize /your/ai/agent.py
```
where `agent.py` is your agent source code. Cognify currently supports unmodified [LangChain](https://github.com/langchain-ai/langchain) and [DSPy](https://github.com/stanfordnlp/dspy) source code. You can also port your existing agent written directly on Python or develop new Python-based workflows with our [simple workflow framework](https://cognify-ai.readthedocs.io/en/latest/user_guide/tutorials/interface/program.html).
Cognify automatically searches for a `config.py` in the same folder as the agent. You can also specify this file explicitly by:
```bash
cognify optimize /your/ai/agent.py -c /your/ai/custom_config.py
```
Within the `config.py`, you should define the following:
- **Sample Dataset**: Cognify relies on training data to evaluate and improve its agent optimization. You should provide a data loader that loads your training dataset in the form of input-output pairs. Read more about how to [load your data](https://cognify-ai.readthedocs.io/en/latest/user_guide/tutorials/dataloader.html).
- **Evaluator**: Cognify expects you to provide an evaluator for judging the final agent generation's quality. To help you get started, Cognify provides several common evaluator implementations such as the F1 score. Find out more about [workflow evaluator](https://cognify-ai.readthedocs.io/en/latest/user_guide/tutorials/evaluator.html).
- **Optimization Configurations and ModelExcerpt of 6,186 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:568ef832525c3c13, topic:agents, topic:agentic-workflow, topic:langgraph