Become a sponsor to Lianmin Zheng
π Hey there! I'm Lianmin Zheng, an open-source enthusiast working on LLMs at UC Berkeley and LMSYS.
Open Source Projects
I am leading or co-leading the development of the following projects with a talented team and a passionate community of contributors.
- FastChat: an open platform for training, serving, and evaluating large language model based chatbots. It powers our LMSYS projects such as Vicuna and Chatbot Arena.
- Vicuna: an open-source chatbot impressing GPT-4 with 90%* ChatGPT quality.
- Chatbot Arena: benchmarking LLMs in the wild with Elo ratings.
- SGLang: an efficient interface and runtime for interaction with LLMs.
- Alpa: an auto-parallelizer for training and serving large-scale neural networks.
Besides, I am fortunate to have made significant contributions to the following projects.
- TVM: an end-to-end deep learning compiler.
- FlexGen: high-throughput generative inference of large language models with a single GPU.
About me
I am a Ph.D. student at UC Berkeley. My research interests are scalable and accelerated systems for machine learning. You can find me on:
π If you are using any of my open-source projects and would like to show your appreciation, you can sponsor me on this page to support the future development of these projects.
Featured work
-
sgl-project/sglang
SGLang is a structured generation language designed for large language models (LLMs). It makes your interaction with models faster and more controllable.
Python 2,604 -
lm-sys/FastChat
An open platform for training, serving, and evaluating large language models. Release repo for Vicuna and Chatbot Arena.
Python 35,147 -
FMInference/FlexGen
Running large language models on a single GPU for throughput-oriented scenarios.
Python 9,047 -
alpa-projects/alpa
Training and serving large-scale neural networks with auto parallelization.
Python 2,998