llm deploy

LLM-Deploy

This tutorial focuses on model/LLM reasoning and deployment theory and practice, and aims to become your partner in mastering the art of LLM reasoning and deployment. Whether you are a newcomer in this field or a veteran looking to deepen your professional skills, you can find the key path to successfully deploying large language models here.

Project Proposal Reasons

Make up for the lack of reasoning and deployment. Provide a good introductory information for more students who are interested in this field or practitioners inside and outside the industry.

Project audience

• Algorithm engineer.
• Students who are interested in reasoning deployment.

Project Highlights

1. Relevant theories and practices are deployed in reasoning.
2. Model and service optimization practice.
3. Comprehensive output from many engineers with practical experience.

Participate in contribution

• If you want to participate in the project, welcome to view the project's Issue to view tasks that are not assigned.
• If you find some issues, please feel free to give feedback in Issue?.
• If you are interested in this project and want to participate, you can communicate through Discussion.

If you are interested in Datawhale and want to launch a new project, please check out the Datawhale contribution guide.

List of contributors

Name	Responsibilities	Introduction	video
Changqin, Yuli	Project leader
Maolin	Chapter 1 Person in charge	Quantification	Link
Yufei	Chapter 2 Person in charge	Distillation
Yuli	Chapter 3: Person in charge	Pruning	Link
Wangyin	Chapter 4: Person in charge	Low rank decomposition
Shu Fan	Chapter 5: Person in charge	express	Link
Spring Sun	Chapter 6: Person in charge	run
Yang Zhuo	Chapter 7: Person in charge	frame
Xue Boyang	Chapter 8 The person in charge	parallel	Link
Jersey Zhang	Chapter 9: Person in charge	concurrent	Link
Li Taiying	Chapter 10 The person in charge	Memory

Note: The header can be customized, but the project manager must be indicated on the list.

Follow us

Scan the QR code below to follow the official account: Datawhale

LICENSE

This work is licensed under the Creative Commons Attribution-Non-Commercial-Share 4.0 International License.

Note: The CC 4.0 protocol is used by default, and other protocols can also be selected according to your own project situation.