Beijing – The open-source DeepSeek-R1 series models have sparked global discussion and attention due to their excellent performance and low development costs. ModelBest, in collaboration with Huawei Ascend, has jointly released the DeepSeek R1 series inference API and cloud image service based on domestic 910B computing power.
Free Experience of DeepSeek R1 Series Inference API
In this release, ModelBest has successfully achieved inference adaptation and optimization of domestic Huawei Ascend 910B computing power and DeepSeek R1 series models based on its self-developed domestic inference engine. The performance is comparable to that of high-end GPUs, providing developers with efficient, flexible, and stable AI inference services, helping companies reduce costs and increase efficiency, and accelerate the rapid deployment of intelligent services based on the domestic software and hardware system.
Empowering Domestic Computing Power, Performance Comparable to High-End GPUs
- Breakthrough Adaptation of Domestic Huawei Ascend Computing Power: ModelBest’s self-developed inference engine has successfully achieved deep adaptation and optimization with domestic Huawei Ascend computing power, benchmarking the inference performance of high-end GPU resources in the industry, and contributing to the construction of the domestic AI ecosystem.
- ModelBest Simultaneously Supports Mainstream Computing Power such as NV H800: In addition to domestic computing power, the DeepSeek R1 series inference API also supports common computing power such as NV H800, fully releasing the potential of heterogeneous clusters and meeting inference needs in different scenarios.
- Flexible Inference Demonstration Model, Covering All Scenarios: The ModelBest cloud platform provides DeepSeek R1 series models with full-blood 671B.
Conclusion:
The collaboration between ModelBest and Huawei Ascend to launch the DeepSeek R1 inference API and cloud image service based on domestic computing power represents a significant step forward in the development of China’s AI ecosystem. By providing free access to powerful inference capabilities, this initiative will empower developers, accelerate the deployment of intelligent applications, and promote the growth of the domestic AI industry. This move not only showcases the increasing capabilities of domestic computing power but also highlights the commitment of Chinese companies to fostering innovation and collaboration in the field of artificial intelligence.
References:
- Machine Heart. (2024, February 4). 免费！潞晨携手华为昇腾，国产算力DeepSeek R1推理API及云镜像服务来了 [Free! ModelBest and Huawei Ascend Launch DeepSeek R1 Inference API and Cloud Image Service Powered by Domestic Computing Power]. Retrieved from [Insert URL if available]
Views: 0