Zhipu GLM-130B
A company incubated by the Department of Computer Science at Tsinghua University, dedicated to creating a new generation of cognitive intelligence general models.
Dark Side of the Moon Kimi
A romantic and mysterious name, produced by Tsinghua University, accessible without a VPN, with nearly 8 million monthly active users, the top-ranked smart chat app in China.
OpenRouter
An initiative to provide a unified UI interface for all LLM models, allowing users to switch between different model chats without registration.
OpenChat
A decentralized smart chat app with the slogan: by the People, for the people; everyone for me, I for everyone.
Yi-34B
An innovative company founded by Dr. Kai-Fu Lee, valued at over $1 billion in less than a year. Its classic model Yi-34B-200K can handle ultra-long contexts, supporting around 300,000 Chinese characters and English words.
Alibaba Tongyi Qianwen
A chatbot released by Alibaba, allowing users to chat with a mobile registration.
Chinese LLM List
A comprehensive collection of LLMs from China with clear sources, including institutions, source information, and classifications.
Hardware Vendors "Selling Shovels"
Nvidia
The first GPU manufacturer to benefit significantly from the AI wave, with a market value that has more than quadrupled.
Huawei
For cutting-edge AI technology in China, Huawei is the go-to, whether in chips or other areas.
Iluvatar
A leading provider of general-purpose high-end GPU chips and supercomputing systems in China.
AI Technology and Software
InfiniBand
This is a networking technology under Nvidia, suitable for high-speed interconnection between high-performance servers, reaching 400 Gbps.
RoCE
RDMA over Converged Ethernet (RoCE), allows remote direct memory access over a single Ethernet.
CUDA
This is a computing platform launched by Nvidia, using a general parallel computing architecture that enables GPUs to solve complex computational problems.
Model Acceleration ColossalAI
Making LLMs run faster, simpler, and cheaper.
Model Inference Triton
A language and compiler for writing efficient custom deep learning primitives used in model inference.
Model Fine-Tuning DataTunerX
A cloud-native model fine-tuning solution using a distributed architecture.
Spiderpool
Spiderpool is an Underlay and RDMA network solution built on Kubernetes.
HwameiStor
A high-performance local storage system suitable for handling stateful workloads and AI workloads.
JuiceFS
A high-performance, cloud-native distributed file system that is elastic, multi-cloud, and compatible with POSIX, HDFS, and S3 protocols.
Domestic AI Pioneer Enterprises
Baichuan Intelligence
This AI startup, established just six months ago, has secured $300 million in funding.
Baidu
Baidu Wenxin Yiyan is the first domestic LLM product charged for C-end users.
Hanwang Technology
A leading company in advanced text recognition technology and intelligent interaction.
iFlytek
The Spark LLM, benchmarking against ChatGPT, is based on deep learning with a focus on Chinese.
Megvii Technology
Focused on machine learning and computer vision technology, excelling in facial recognition, image recognition, and intelligent monitoring.
SenseTime
Covering key technology fields such as perceptual intelligence, natural language processing, decision intelligence, and intelligent content generation.
AISpeech
A company focused on natural language processing and voice interaction technology.
:simple-tencentqq:{ .lg .middle } Tencent
Performing well in natural language processing, computer vision, and speech recognition, widely applied in gaming, social networking, and smart finance.
CloudWalk Technology
A company focused on human-computer interaction technology and artificial intelligence applications.
UniSound
An innovative enterprise focused on smart IoT services, covering the research and manufacturing of systems like speech recognition and semantic understanding.
ByteDance
ByteDance is secretly developing multiple products in the AI LLM field, including multimodal digital human products and AI-generated images and videos.