Simple, affordable large model training and serving

Alpa is a system for training and serving gigantic machine learning models.
Alpa makes training and serving large models like GPT-3 simple, cost-effective, accessible to everyone.

Free, Unlimited OPT-175B Text Generation

Examples: Fact Chatbot Philosophy Translation Cryptocurrency Elon Musk Math

64
0.7
0.5
Please be patient. Your generation may take at most X seconds.

Frequently Asked Questions

Alpa is an open-source system for training large-scale neural networks. Alpa aims to automate large-scale distributed training and serving with just a few lines of code. Alpa was initially developed by folks in the Sky Lab, UC Berkeley. Some advanced techniques used in Alpa have been written in a paper published in OSDI'2022. Alpa community is growing with new contributors from Google, Amazon, and Meta.

OPT-175B is a GPT-3 equivalent model trained by Meta and made publicly available in the MetaSeq project. It is by far the largest pretrained language model available in open source, with 175 billion parameters. For detailed performance of OPT-175B, check the OPT paper.

You can start with the provided examples. Avoid spaces at the end of your query. New lines are great though. More examples can be found in the appendix of the OPT paper. WARNING: This model will generate many offensive things. Due to this being an alpha demo, no safety measures are in place.

Because Alpa is more automatic, scalable, and cost-effective.

It depends on which types of GPUs used. A hard constraint now is that the total GPU memory in the cluster needs to be greater than 350GB in order to successfully run the model inference. Many existing training or serving systems usually rely on using the latest generations of GPUs with the largest memory capacity, such as 80GB A100. In contrast, Alpa, due to its more powerful backend, enables serving OPT-175B with more flexible parallelisms on older generations of GPUs, such as 40GB A100, V100, T4, M60, etc.
Take an example, if you choose to use 16GB V100 GPUs, then you would need 350 / 16 = 22 V100 GPUs to run the service.

Alpa does not require the latest generation GPUs (such as 80GB A100 to do the work), hence reduces the machine cost. With that, we leverage older generations of hardware provided by our sponsors: MBZUAI and Sky Lab, UC Berkeley.

Your usage of this service is subject to Alpa's open source license. Your usage of the OPT-175B model is subject to Meta's OPT-175B license.

Interested in contributing to the Alpa project?