I'm Li Yin ([GitHub](https://github.com/liyin2015)), the author of AdalFlow and a former AI researcher at Meta AI.
AdalFlow was inspired by a viral [LinkedIn post](https://www.linkedin.com/posts/li-yin-ai_both-ai-research-an...) I made, discussing how the LLM ecosystem lacks a shared library that bridges the gap between research and product development—similar to how PyTorch has streamlined model training and adaptation.
I decided to build this library while working on my product, a conversational search engine called [Sylph](https://sylph.ai/). After trying out existing libraries and finding that I had to write everything myself, I ended up with a solution that was lighter, faster, and offered more control. However, managing the codebase soon became overwhelming.
AdalFlow is based on my vision for the future of LLM applications, which I see as a three-stage workflow:
- *V1*: Use the library to quickly build your initial task pipeline, getting you 70-80% of the way to production.
- *V2*: Auto-optimize the prompt to push an additional 10%, bringing your product to a near-ready state without the hassle of manual prompt iteration.
- *V3*: Leverage V2 to label more data. As more users interact with your product, the next step is to fine-tune the LLM, further optimizing for speed, accuracy, and cost-effectiveness.
We've completed V1 and V2. Our auto-optimizer can enhance GPT-3.5 performance to match that of GPT-4, making any task nearly production-ready. Our architecture is the most robust, lightweight, and modular, with our auto-optimizer being the most accurate—even when compared to Dspy and Text-Grad. We have three research papers coming out soon that will explain how we achieved this. This is the first time the library has been released ahead of the research papers.
It’s definitely worth checking out—you might be surprised by the results. We've had similar experiences using PyTorch and PyTorch Lightning.
Thanks for the insightful response. Good point on using 4o-mini to save cost. I'll try it out.
I will check more into the soft-prompt tuning.
For the current scope, we are focused on in-context learning, ways to improve model reasoning at the inference time.
We use auto-differentiative framework (backpropagation) to do zero-shot instruction optimization and few-shot demonstration. currently even just zero-shot can often surpass Dspy's few-shots (as many as 40 shots). And I have come up a training paradigm that will (1) start zero-shot (2) review performance from advanced teacher model to see if we can have a gap to gain from the teacher. (3) if there is a gap to teacher, we start to do low-shot demonstrations, and gradually increase the number of shots.
AdalFlow is named in honor of Ada Lovelace, the pioneering female mathematician who first recognized that machines could do more than just calculations. As a female-led team, we aim to inspire more women to enter the AI field.
LLM applications are messy, but AdalFlow has made it elegant!
0.2.0 release highlight a unified auto-differentiative framework where you can perform both instruction and few-shot optimization. Along with our own research, “Learn-to-Reason Few-shot In-context Learning” and “Text-Grad 2.0”, AdalFlow optimizer converge faster, more token efficient, and with better accuracy than optimization-focused frameworks like Dspy and text-grad.
Is AdalFlow also focused on automated prompt optimization or is it broader in scope? It looks like there are also some features around evaluation. I'd be really interested to see a comparison between AdalFlow, DSPy [0], LangChain [1] and magentic [2] (package I've created, narrower in scope).
We are broader. We have essential building blocks for RAG, Agents. But also made whatever you build possible to auto-optimize. You can think of us as the library to do in-context learning. Just like PyTorch is for model-training.
We have better accuracy, more token-efficient, and faster convergence speed. We are publishing three research papers to explain this better to researchers.
We will compare with these optimization libraries but wont compare with libraries like LangChain or LlamaIndex. As they simply dont have optimization and it is pain to build on them.
Thanks for the explanation! Do you see auto-optimization as something that is useful for every use case or just some? And what determines when this is useful vs not?
I would say its useful for all production-grad application.
Trainer.diagnose helps you get a final eval score across different splits of datasets: train, val, test, and it logs all errors, including format errors so that you can manually diagnose and to decide if the evaluation is too low that you need further text-grad optimization.
if there is still a big gap between your optimized prompt vs performance on a more advanced model with the same prompt (say gpt4o), then you can use our "Learn-to-reason few-shot" to create demonstration from the advanced model to further close the performance gap. We have use cases optimized the performance all the way from 60% to 94% on gpt3.5 and the gpt4o has 98%.
We will give users some guideline in general.
We are the only library provides "diagnose" and "debug" feature and a clear optimization goal.
Folks, let's focus on the value instead of just bashing their name and comparing it to PyTorch. The team has an AI background, so naturally, PyTorch feels like home to them. Given the state of the existing libraries, it is great to see another one taking a completely light approach.
LightRAG follows three fundamental principles from day one: simplicity over complexity, quality over quantity, and optimizing over building. This design philosophy results in a library with bare minimum abstraction, providing developers with maximum customizability. View Class hierarchy here. https://lightrag.sylph.ai/developer_notes/class_hierarchy.ht...
Now helping founders automate investor reach out flow. Our search performs better than structured search like NFX signal, and conversational search like Perplexity AI and Google gemini.
Now it private testing. first 50 users will be able to try it first!
Hi! I’m Li Yin, and I’m super excited to launch SylphAI to the community. SylphAI is your go-to place to either hire an AI team to deliver a pressing AI project on-hand or hire a top expert to advise on your AI project setup.
The Problem — For big companies, such as Google, Amazon, Meta, roughly 50% are contractors. But the quality of AI contractors is low. For small companies, they either lack money or that the brand is unable to attract top AI talent yet. There is a need for on-demand and high-quality AI workforce. The existing solution such as freelancers or consulting firms is both low-quality and high-cost in the AI space.
* Neither the existing generic freelancer platform nor consulting firms are known for attaining top AI talent
* Most AI projects are too complicated or the scope is too big for a single freelancer to complete.
Managing a few freelancers/contractors could potentially waste a lot of full-time employees' time.
* Consulting firms have too much operational cost and profit margin, thus the pay that finally got into the hands of AI talent is small.
* Additionally, there is a shortage of AI talent, especially those with hands-on productionalization experience.
Our Solution — We allow AI freelancers to form teams similar to what you see at any company in the hiring, managing, and operation process. To address challenging projects we would staff a tech lead who is immersed on that topic to lead and plan with less work hours, and more engineers who might be slightly less experienced to be on the implementation side with more work hours. In all, we deliver effective outcomes in both quality and cost.
Since the launching of our landing page (sylphai.com), we have received high enthusiasm from both AI talent and clients. We have more than 60 AI engineers/researchers signed up and many have a background in top companies and universities. We have started to work with a bay area startup using this on-demand AI team structure, and the quality of the work is high.
We envision a future where a company's AI workforce consists of half full-time AI teams and half on-demand AI teams, working in harmony and effectively together to deliver the best outcome.
Our asks —
* If your business or any of your acquaintance needs an on-demand AI team or top expert consulting, shoot us a message with your requirements or give us an intro. We offer 30 min free generic AI consulting and a discounted rate. * Also, we would love your feedback on our landing page and value proposition.
* If you are an AI engineer, we would love to hear your thoughts. Half of the core of SylphAI is to provide all AI talent an equal alternative to full-time corporate life.
* We are also preparing fundraising, if you know any angels or VCs who can potentially have interest with us, would appreciate it a lot if you give us a warm intro.
* If you are just in general interested with us, shoot us a message too. Would love to connect and get some feedback.
I'm Li Yin ([GitHub](https://github.com/liyin2015)), the author of AdalFlow and a former AI researcher at Meta AI.
AdalFlow was inspired by a viral [LinkedIn post](https://www.linkedin.com/posts/li-yin-ai_both-ai-research-an...) I made, discussing how the LLM ecosystem lacks a shared library that bridges the gap between research and product development—similar to how PyTorch has streamlined model training and adaptation.
I decided to build this library while working on my product, a conversational search engine called [Sylph](https://sylph.ai/). After trying out existing libraries and finding that I had to write everything myself, I ended up with a solution that was lighter, faster, and offered more control. However, managing the codebase soon became overwhelming.
AdalFlow is based on my vision for the future of LLM applications, which I see as a three-stage workflow:
- *V1*: Use the library to quickly build your initial task pipeline, getting you 70-80% of the way to production. - *V2*: Auto-optimize the prompt to push an additional 10%, bringing your product to a near-ready state without the hassle of manual prompt iteration. - *V3*: Leverage V2 to label more data. As more users interact with your product, the next step is to fine-tune the LLM, further optimizing for speed, accuracy, and cost-effectiveness.
We've completed V1 and V2. Our auto-optimizer can enhance GPT-3.5 performance to match that of GPT-4, making any task nearly production-ready. Our architecture is the most robust, lightweight, and modular, with our auto-optimizer being the most accurate—even when compared to Dspy and Text-Grad. We have three research papers coming out soon that will explain how we achieved this. This is the first time the library has been released ahead of the research papers.
It’s definitely worth checking out—you might be surprised by the results. We've had similar experiences using PyTorch and PyTorch Lightning.
To learn more about our optimizer, visit: https://adalflow.sylph.ai/use_cases/classification.html.
Best,
Li