A knowledge-focused large language model platform from CNKI and Huawei, built for authoritative content, research support, and professional question answering.
OrionStar's Orion-14B is an enterprise-focused large language model with 32K token context, quantization, and multilingual support for long-text processing.
孟子大模型 is Langboat's self-developed large pre-trained language model, built for multilingual, multimodal understanding and generation across industry use cases.
ModelBest (面壁智能) builds the CPM family of pretrained language models and the ModelForce platform for efficient training, fine-tuning, compression and inference.
RockAI's enterprise foundation model platform offers efficient inference, fast pretraining, knowledge management and API-based integration for business workflows.
盛典AI is a Chinese enterprise AI think-tank service that supports digital transformation, data analysis and decision-making for small and midsize businesses.
Huazang is Xiaoi's self-developed general-purpose foundation model, built for customizable, controllable enterprise applications such as customer service, analytics, and decision support.
RobotGPT is 达闼's multimodal foundation model built for robots, combining language, image understanding and behavior generation across industry scenarios.
Mobvoi's Sequence Monkey open platform delivers multilingual language understanding, generation, Q&A, and code assistance through APIs and tailored enterprise deployment.
Caozhi is Datagrand's Chinese vertical-domain large language model for drafting long documents, reports, and application materials with guided writing.
Baichuan is a family of bilingual large language models from Baichuan Intelligence, offering long-context reasoning, search augmentation and open weights for commercial use.
甲子光年 (Jiaziguangnian) is a China-based technology think tank and media platform focused on AI, hard tech, and the commercialization of emerging technologies.
Huawei Atlas 900 is a powerful AI training cluster built with Ascend 910 processors, delivering up to 1024 PFLOPS for research and enterprise AI workloads.