At the recent Zhongcheng Hualong 2026 GPU product launch conference, Zhongcheng Hualong officially launched the new HL200 inference chip and supernode intelligent computing cluster solution to achieve a leapfrog upgrade of domestic AI inference computing power from a single-point chip breakthrough to a 10,000-card cluster systematic collaboration. Bi Kaichun, Executive Deputy Director and Secretary General of the Electronic Information Technology Commission of the Ministry of Industry and Information Technology, and Liu Zhiguang, Chairman of the China Science and Technology Consulting Association, attended the event and delivered speeches. More than 100 representatives from national authorities, research institutes, industrial investment institutions, central state-owned enterprises and core partners in the industrial chain attended the conference. Facing the commercial implementation of the trillion-parameter model, Zhongcheng Hualong launched the second-generation HL200 inference chip, which is suitable for generative AI and AI agent high concurrency inference scenarios. The HL200 natively supports FP4 and FP8 ultra-low precision inference, is compatible with mainstream FP16 accuracy, and covers the computational power requirements of large models in all dimensions. According to the data, the FP16/BF16 single chip card has 0.5P computing power, FP8 computing power 2P, and FP4 computing power 4P, with an energy efficiency ratio of 5.12 TFLOPS/W, which is stable in the first tier in China, effectively solving the pain points of low energy efficiency and high deployment costs in the industry. At the ecological level, HL200 is fully compatible with mainstream technology stacks such as PyTorch, ONNX, and VLLM, and is equipped with OpenAI compatible interfaces and lightweight migration solutions to greatly reduce enterprise model migration and business switching costs.

At the recent Zhongcheng Hualong 2026 GPU product launch conference, Zhongcheng Hualong officially launched the new HL200 inference chip and supernode intelligent computing cluster solution to achieve a leapfrog upgrade of domestic AI inference computing power from a single-point chip breakthrough to a 10,000-card cluster systematic collaboration. Bi Kaichun, Executive Deputy Director and Secretary General of the Electronic Information Technology Commission of the Ministry of Industry and Information Technology, and Liu Zhiguang, Chairman of the China Science and Technology Consulting Association, attended the event and delivered speeches. More than 100 representatives from national authorities, research institutes, industrial investment institutions, central state-owned enterprises and core partners in the industrial chain attended the conference. Facing the commercial implementation of the trillion-parameter model, Zhongcheng Hualong launched the second-generation HL200 inference chip, which is suitable for generative AI and AI agent high concurrency inference scenarios. The HL200 natively supports FP4 and FP8 ultra-low precision inference, is compatible with mainstream FP16 accuracy, and covers the computational power requirements of large models in all dimensions. According to the data, the FP16/BF16 single chip card has 0.5P computing power, FP8 computing power 2P, and FP4 computing power 4P, with an energy efficiency ratio of 5.12 TFLOPS/W, which is stable in the first tier in China, effectively solving the pain points of low energy efficiency and high deployment costs in the industry. At the ecological level, HL200 is fully compatible with mainstream technology stacks such as PyTorch, ONNX, and VLLM, and is equipped with OpenAI compatible interfaces and lightweight migration solutions to greatly reduce enterprise model migration and business switching costs.